EFFECT OF SOLAR PV ON TRANSIENT STABILITY OF THE NEW ZEALAND POWER SYSTEM

Transcription

1 EFFECT OF SOLAR PV ON TRANSIENT STABILITY OF THE NEW ZEALAND POWER SYSTEM DECEMBER 2017 TECHNICAL REPORT

2

3 Table of Contents Table of Contents EXECUTIVE SUMMARY... VIII 1 INTRODUCTION Programme Overview The PV Generation Investigation Project The New Zealand power system Impacts on Transient Stability TRANSIENT STABILITY CRITERIA AND PERFORMANCE REQUIREMENTS Definitions Factors influencing transient stability Criteria used in New Zealand ASSUMPTIONS AND STUDY METHODOLOGY Study assumptions Study inputs Study methodology EFFECT ON TRANSIENT STABILITY Transmission line loading Three phase-to-ground fault at Wairakei 220 kv bus followed by disconnection of Whakamaru- Wairakei-1 circuit from service Three phase-to-ground fault at Huntly 220 kv bus followed by disconnection of Huntly-Stratford- 1 line from service Three phase-to-ground fault at Clyde 220 kv bus followed by disconnection of Clyde-Twizel-1 circuit from service Three phase-to-ground fault at Manapouri 220 kv bus followed by disconnection of Manapouri- North Makarewa-1 and 2 circuits North Island other scenarios South Island - other scenarios KEY FINDINGS AND CONCLUSIONS Transient stability of the power system Effect of low inertia RECOMMENDATIONS Future work A1 CASE DETAILS A1.1 ON/OFF status of North Island conventional generators A1.2 ON/OFF Status of South Island conventional generators A1.3 MW and Mvar output of North Island conventional generators A1.4 MW and Mvar output of South Island conventional generators A2 THE PV MODELS USED A3 TRANSIENT RESPONSES A3.1 Rotor angle of North Island generators A3.2 Rotor angle of South Island generators A4 PV INVERTER DYNAMIC MODEL RESPONSE A4.1 Frequency response A4.2 Transmission voltage and reactive power response Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. iii

4 Table of Contents A4.3 Reconnecting characteristics A5 TRANSIENT STABILITY THEORY A6 EMERGING ENERGY PROGRAMME: PLAN AND OUTCOME STRATEGY A6.1 Emerging Energy Technologies Outcome Strategy Map GLOSSARY OF TERMS AND ACRONYMS BIBLIOGRAPHY Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. iv

5 Table of Figures Table of Figures Figure 1: 220 kv bus faults simulated in North Island... 9 Figure 2: 220 kv bus faults simulated in South Island Figure 3: Rotor angle response with first swing instability Figure 4: Rotor angle response with oscillatory instability Figure 5: Winter Tuesday 27/07/2015 sunny day scenarios Figure 6: Winter Saturday 01/08/2015 sunny day scenarios Figure 7: Summer Sunday 10/01/2016 sunny day scenarios Figure 8: Installed PV inverter types as a percentage of total Grid Zone PV inverters Figure 9: Grid Zone PV penetration levels (winter, Tuesday) Figure 10: Example of a PV inverter grid connection configuration in the power-flow Figure 11: Process used to allocate MW output for each inverter at a given grid exit point Figure 12: active power fault response of inverter type A, B and C Figure 13: Inverter RMS current output for a voltage disturbance step signal Figure 14: High level overview of the transient stability assessment process Figure 15: Loadings on the transmission lines in upper North Island region in winter base cases Figure 16: Loadings on the transmission lines in upper South Island region in Winter base cases Figure 17: Loadings on transmission circuits in upper North Island region in Summer base cases Figure 18: Loadings on transmission circuit in upper South Island region in Summer base cases Figure 19: Portion of the 22kV network in Bay of Plenty region Figure 20: Rotor angle of Bay Wairakei G1 generating unit. Ref gen. Glenbrook G1 (winter Tuesday) Figure 21: rotor angle of Matahina G1 generating unit. Ref. gen Glenbrook G1 (winter Saturday) Figure 22: rotor angle of Te Mihi G1 generator. Ref. gen Glenbrook G1 (summer Sunday) Figure 23: North Island total PV MW output response monitored as an interface Figure 24: Wairakei G1 machine speed (winter Tuesday) Figure 25: Wairakei station generation (winter Tuesday) Figure 26: Transient stability index for various δmax values Figure 27: Transient stability indexes for the Wairakei 220 kv fault (winter Tuesday) Figure 28: Transient stability indexes for the Wairakei 220 kv fault (winter Saturday) Figure 29: Transient stability indexes for Wairakei 220 kv fault (summer Sunday) Figure 30: Portion of the 220 kv network between Taranaki and Auckland Figure 31: Rotor angle Mokai G1 generator. Ref. gen. Glenbrook G1 (winter Tuesday) Figure 32: Rotor angle Wairakei G1 generating unit. Ref. gen Glenbrook G1 (winter Saturday) Figure 33: Rotor angle Te Mihi G1 generator. Reg. gen. Glenbrook (summer Sunday) Figure 34: Maraetai G2 machines speed (winter Tuesday) Figure 35: Maraetai station generation (winter Tuesday) Figure 36: North Island PV inverter MW response (winter Tuesday) Figure 37: North Island PV inverter MW response (winter Saturday) Figure 38: North Island PV Inverter MW response (summer Sunday) Figure 39: Transient stability indexes for the Huntly 220 kv fault (winter Tuesday) Figure 40: Transient stability indexes for the Huntly 220 kv fault (winter Saturday) Figure 41: Transient stability indexes for the Huntly 220 kv fault (summer Sunday) Figure 42: Portion of the 22kV network in Southland and Otago Figure 43: Rotor angle of Clyde G1. Ref. gen. Benmore G1 (winter Tuesday) Figure 44: Rotor angle of Roxburgh G1, Ref. gen. Benmore G1 (winter Tuesday) Figure 45: Rotor angle of Ohau A- G4. Ref. gen. Benmore G1 (winter Tuesday) Figure 46: Rotor angle of Ohau B- G10 (winter Tuesday) Figure 47: Clyde-Twizel-1 & 2 MW flow, Clyde station MW and num. of Clyde units online (winter Tuesday) Figure 48: Bus voltage at Clyde 220 kv faulted bus Figure 49: Machine speed of Clyde G1 generating unit (winter Tuesday) Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. v

6 Table of Figures Figure 50: Transient stability indexes for the Clyde 220 kv fault (winter Tuesday) Figure 51: Transient stability indexes for the Clyde 220 kv fault (winter Saturday) Figure 52: Transient stability indexes for the Clyde 220 kv fault (summer Sunday) Figure 53: Lower South Island 220 kv network Figure 54: Rotor angle Manapouri G1 (winter Tuesday) Figure 55: Machine speed Manapouri G1 (winter Tuesday) Figure 56: Rotor angle Manapouri G1 (winter Saturday) Figure 57: Machine speed Manapouri G1 (winter Saturday) Figure 58: Machine speed of Argyle G1 (winter Saturday) Figure 59: Rotor angle Manapouri G1 (summer Sunday) Figure 60: Machine speed of Manapouri G1 (summer Sunday) Figure 61: Transient stability indexes for the Manapouri 220 kv fault (winter Tuesday) Figure 62: Transient stability indexes for the Manapouri 220 kv fault (winter Saturday) Figure 63: Transient stability indexes for Manapouri 220 kv fault (summer Sunday) Figure 64: Rotor angle of Mokai G1 for various faults (winter Tuesday, case 5) Figure 65: Transient stability indexes of various fault scenarios in North Island (winter Tuesday) Figure 66: Rotor angle of Roxburgh G1 unit for SCN23 (winter Tuesday) Figure 67: Transient stability indexes of various fault scenarios in South Island (winter Saturday) Figure 68: Solar PV uptake in New Zealand to May 2017 [11] Figure 69: Frequency control component of inverter type C dynamic model (simplified) Figure 70: Current limiting logic modelled for inverter type A Figure 71: Simplified dynamic model control block diagram of Volt-Var component of inverter type A Figure 72: Simplified dynamic model control block diagram of Power Factor component of inverter type A. 76 Figure 73: Dynamic model control block diagram of voltage tripping, frequency tripping and reconnect settings component of inverter type C Figure 74: Rotor angle response of North Island generators. Ref gen. Glenbrook G1 (winter Tuesday, Wairakei fault) Figure 75: Rotor angle response of North Island generators. Ref gen. Glenbrook G1 (winter Saturday, Wairakei fault) Figure 76: Rotor angle response of North Island generators. Ref gen. Glenbrook G1 (summer Sunday, Wairakei fault) Figure 77: Rotor angle response of North Island generators. Ref gen. Glenbrook G1 (winter Tuesday, Huntly fault) Figure 78: Rotor angle response of North Island generators. Ref gen. Glenbrook G1 (winter Saturday, Huntly fault) Figure 79: Rotor angle response of North Island generators. Ref gen. Glenbrook G1 (summer Sunday, Huntly fault) Figure 80: Rotor angle response of South Island generators. Ref gen. Benmore G1 (winter Tuesday, Clyde fault) Figure 81: Rotor angle response of South Island generators. Ref gen. Benmore G1 (winter Saturday, Clyde fault) Figure 82: Rotor angle response of South Island generators. Ref gen. Benmore G1 (summer Sunday, Clyde fault) Figure 83: Rotor angle response of South Island generators. Ref gen. Benmore G1 (winter Tuesday, Manapouri fault) Figure 84: Rotor angle response of South Island generators. Ref gen. Benmore G1 (winter Saturday, Manapouri fault) Figure 85: Rotor angle response of South Island generators. Ref gen. Benmore G1 (summer Sunday, Manapouri fault) Figure 86: Inverter over-frequency response Figure 87: Volt-Var response of inverter type A (voltage measured at the LV side of the inverter transformer winding) Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. vi

7 Table of Figures Figure 88: Constant Power Factor response of inverter type A Figure 89: Reconnecting characteristics of inverter types A, B and C for voltage dip in terminal voltage Figure 90: Two generator network with reactance between them [3] Figure 91: Power-flow between two buses based on the angle difference between them [3] Table 1: North Island bus faults... 8 Table 2: South Island bus faults... 8 Table 3: Demand shedding to maintain transient stability of New Zealand Table 4: PV Inverter types Table 5: PV inverter type allocations Table 6: Power-flow scenarios used for the transient stability study Table 7: Inverter voltage and frequency ride trip settings used in the dynamic models Table 8: Transient stability study scenarios Table 9: North Island PV interface MW flow (winter Tuesday) Table 10: Statuses of North Island conventional generators in various power-flow cases used (MW max descending order) Table 11: Statuses of South Island conventional generators in various power-flow cases used (MW max descending order) Table 12: MW and MVAR output of North Island generation in the power-flow cases Table 13: MW and MVAR output of South Island generation in the power-flow cases Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. vii

8 Executive Summary EXECUTIVE SUMMARY Transpower has initiated a programme of work to investigate the impacts on the power system from an anticipated increase in distributed, non-dispatchable and renewable generation, and from other emerging technologies in New Zealand. The aim of Transpower s Emerging Energy Programme is to identify potential compromise to Transpower s ability to meet the system operator Principal Performance Obligations (PPOs) with the introduction of new generation technologies. The alternative would involve a revision of the PPOs to accommodate these emerging technologies. The first part of the programme features the PV Generation Investigation Project, which studies the effect of PV generation technology on four areas of the power system: generation dispatch, frequency management, transmission voltage management and transient stability. Study reports have been produced for each of these areas, with this report covering the study into the likelihood of high PV generation levels causing first swing transient stability issues. All the project studies used a scenario of 4 GW [1] of solar PV capacity installed, as discussed in the generation dispatch study (refer to Effect of Solar PV on Generation Dispatch in New Zealand). It should be noted that the assumptions detailed in section 3 of this transient stability report are important to this study, particularly the fault ride-through assumption for the new PV generation. All conclusions, observations and discussions in this report depend on the appropriateness of these assumptions. The study described in this report concludes that: 1. First swing transient instability is unlikely to occur even with high PV penetration levels in the New Zealand power system. 2. The addition of PV generation is expected to result in a reduction in system inertia and although this may have a detrimental effect on transient stability, the study shows this to be acceptable, as the remaining conventional generation remains stable. The study shows there are several factors related to PV generation that can improve the transient stability of the New Zealand power system: 1. PV generation is connected close to load centres, which reduces power flows in the grid. 2. PV generation itself is inherently stable and not prone to participate in any generator or system oscillations. 3. PV generation is displacing the larger conventional generating units so these generators are not connected when the system has low inertia. These factors are discussed in detail in this report. Several recommendations are made for future studies to expand on the initial work presented here. These include: Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. viii

9 Executive Summary 1. A requirement for more detailed modelling, as PV generation in New Zealand increases. 2. Undertaking continued assessment into the impact of increased PV generation on the transient stability of the power system. The learnings gained in these studies will be useful in steering the future of the system operator service, electricity market design, industry regulations, policies and procedures for a period of increasingly decentralised supply and responsive consumer technologies. Ultimately, this understanding will facilitate an evolving power system which can continue to meet the changing needs of New Zealanders. Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. ix

10 Section 1 Introduction 1 INTRODUCTION 1.1 Programme Overview Transpower's Emerging Energy Programme investigates the potential impacts on the power system resulting from an anticipated increase in distributed, non-dispatchable, renewable generation and other emerging technologies in New Zealand. The programme outlines the strategy Transpower has adopted to develop its capability and business processes to enable a successful integration of such technologies in the New Zealand power system. See Appendix A6 for a scope of work flowchart that summarises Transpower's Emerging Energy Programme The growth of distributed, non-dispatchable renewable generation Distributed, non-dispatchable electricity generation, primarily PV generation, has grown rapidly in most regions around the world in recent years. The change in technology costs, consumer preferences, policies and environmental concerns, leads to this trend of growth [1]. PV generation uptake is still relatively low in New Zealand. However, the rate of growth is expected to increase for the foreseeable future, with PV generation projected to become a significant part of New Zealand's electricity supply mix. Other emerging technologies (such as energy storage devices, Home Energy Management Systems (HEMS), Electric Vehicles (EVs) and smart appliances) will also play a role in shaping the future of the New Zealand power system. New business models for energy trading and distributed generation ownership will facilitate consumer choice and change the way we produce and use electricity. Though the cumulative effect of these developments is highly interdependent and difficult to predict, the electricity industry will need to be proactive in meeting changing consumer expectations and a shifting market environment, to avoid significant business disruptions Assessing New Zealand's ability to adapt to new technologies The New Zealand power system has some unique features not the least of which is being an islanded system with a high proportion of electricity generated from hydro-power backed by storage. A 2008 study of the system's ability to accommodate wind generation indicated that hydro generation afforded a high degree of flexibility to accommodate intermittent generation. Transpower is assessing the possible future impacts of intermittent generation technologies on the power system and the policies it may need to adopt to continue to meet the PPOs in its role as system operator. These assessments will also provide useful context for the future development of the Electricity Industry Participation Code (the Code), including the PPOs The challenges due to the variability of PV Generation Electricity produced from solar irradiance depends on the position of the sun, which is predictable though variable. With consistently clear or overcast weather, PV generation output can be relatively steady; with output increasing in the early morning and Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 1

11 Section 1 Introduction decreasing during the late afternoon. However, PV generation output can be highly variable with changeable and fast moving cloud. The variability and intermittent effects of PV generation can cause operational issues for grid management. The increase in inverter-based generation in the power system (replacing conventional synchronous generators) can alter the dynamic behaviour of the power system. Inverters are highly programmable making their behaviour less predictable. Furthermore, PV generation will be more distributed compared to the present centralised generation topology. This form of distributed generation presents challenges in studying the dynamic behaviour and real-time operation of the power system. However, the studies are needed in order to understand the effect of PV generation variability and intermittency on the power system, and in forecasting the likely impact on the reliability of the ancillary services. 1.2 The PV Generation Investigation Project Transpower's PV Generation Investigation Project is part of the wider Emerging Energy Programme to ensure a smooth integration of these new technologies in New Zealand. The PV Generation Investigation Project provides studies into PV generation technologies and can be broadly separated into four main areas: generation dispatch, frequency management, transmission voltage management and transient stability. Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 2

12 Section 1 Introduction 1.3 The New Zealand power system Overview The New Zealand power system has several features which have the potential to impact the integration of distributed, non-dispatchable generation. The major factor is that ours is an isolated system with a high proportion of electricity generated from renewable sources which can vary in availability; namely hydro and wind generation. It is necessary to understand the impact to New Zealand's security of supply due to additional variable energy sources that are not highly correlated to either hydrology or wind resources. A significant increase in the share of PV generation in the generation mix may require changes to the existing transmission network equipment, operational processes, code and industry standards to: Secure adequate responsive generation (and possibly energy storage) capacity to manage the variable and intermittent nature of non-dispatchable PV generation. Introduce new equipment and operational measures to ensure adequate grid stability and control. Include distributed PV generation forecasting into scheduling processes. Ensure prices reflect economic costs. In reading the study reports produced for the PV Generation Investigation Project it is assumed the reader is familiar with the New Zealand power system, including the following key features: There is good generation mix with approximately 80% of electricity supply from variable renewable sources. There is existing thermal plant. There have been recent thermal plant retirements. There is existing wind penetration. It is a two island system; it is relatively small, with low inertia at times and large generating units present susceptibility to frequency disturbances. There is a mix of generation characteristics - fast ramping hydro, slower ramping thermal, constant geothermal, variable wind, etc. Transpower holds a classification of power system risks. Ancillary services are used to manage frequency - FIR, SIR, IL, FK, AUFLS (see the glossary for details of these) PV generation in New Zealand PV generation uptake in New Zealand has been relatively low. As of December 2017, New Zealand's installed PV generation capacity has grown to a total of 62 MW, with generation from residential, commercial and industrial sites [1]. This growth places total installed PV Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 3

13 Section 1 Introduction generation capacity at a level similar to the smaller, run-of-river hydro stations in New Zealand. However, at typical New Zealand solar capacity factors, this installed generation supplies only around 0.1% of total national energy consumption. This level of PV generation capacity has not compromised our ability to operate the power system securely and economically, with the existing tools and policies. However, the rate of growth is rapid, with a doubling time for installed PV generation capacity of approximately 18 months. PV generation installations are expected to continue to grow, as falling costs and an expanding market drive an increasing pace of PV generation uptake. Integration of high levels of PV generation into the power system will impact the frequency response to system imbalance, for the reasons outlined below: It is distributed and non-dispatchable, and therefore offsets load behind the GXP, with limited system operator visibility. It is highly stochastic, with rapid changes in output possible, depending on the relevant temporal and spatial scales, type of weather, season and level of uptake. The normal PV generation profile is negatively correlated with demand at times of maximum peak PV generation during mid-day, resulting in low system inertia that is susceptible to frequency disturbance. Inverter-based PV generation exhibits different frequency behaviour when subjected to system imbalance compared to conventional synchronous generation. 1.4 Impacts on Transient Stability This report discusses the study done to determine the likely effect of increasing PV generation on the transient stability of the New Zealand power system. Transient stability is the terminology used to describe how the power system responds to a sudden disturbance such as a line fault. This is described in more technical detail in section 2. Reliability of the power system depends on its ability to ride through a disturbance (e.g., loss of a generator, loss of a transmission circuit, loss of a large load, occurrence of a system fault) and remain in a stable operating state. The power system is expected to provide continuous electricity supply with voltage and frequency operating within the statutory ranges. Roof-top PV generation is a form of distributed and non-dispatchable generation with a very small unit size. An increase in PV generation may displace grid-connected synchronous generation, which would change the generation mix and alter the pattern of power-flowing through the transmission network. This could result in significant changes to the way Transpower plans and operates the power system. The transient stability of a generator connected to a conventional power system relies on the interaction of the generator with the power system, including the loads and, importantly, with any large generators connected at transmission voltages. Large Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 4

14 Section 1 Introduction transmission connected generators result in a dynamic re synchronising force which acts to keep each generator connected to the grid and synchronised with other generators. The impact of distributed generation made up of small generation blocks is that it significantly changes this dynamic behaviour. Reducing the number of large generators has been reported as reducing the stability of the remaining transmission-connected generators in several power systems around the world [2]. This study considers the likely effects on the New Zealand power system. Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 5

15 Section 2 Transient Stability Criteria and Performance Requirements 2 TRANSIENT STABILITY CRITERIA AND PERFORMANCE REQUIREMENTS 2.1 Definitions Transients are disturbances that occur for a short duration and when the disturbance ends or is removed by automatic switching, the power system returns to steady state operation. Transients are caused by normal switching events or external events such as faults. System faults have been used to assess transient stability of the New Zealand power system in this work. The literature describes transient stability or large-disturbance rotor angle stability as the ability of the power system to maintain synchronous operation when subjected to a severe disturbance, such as a short circuit on a transmission line. The resulting system response involves large excursions of generator rotor angles and is influenced by the nonlinear power-angle relationships [3]. Instability that may result occurs in the form of increasing rotor angle of one or more synchronous generating units relative to the rotor angles of other generating units, leading to their loss of synchronism with those synchronous generating units. There are three common causes of the loss of synchronism: insufficient synchronising torque, insufficient damping torque and protection system operation due to the magnitude of a power swing. Insufficient damping torque is commonly referred to as a small-signal stability problem, but it can often be observed in transient stability studies. At present, the most practical available method of transient stability analysis is time-domain simulation in which the nonlinear differential equations are solved by using step-by-step numerical integration techniques. 2.2 Factors influencing transient stability Transient stability depends on both the initial operating state of the system and the severity of the disturbance. Instability is usually due to insufficient synchronising torque, manifesting as first swing instability. Nevertheless, in large power systems, transient stability may not always occur as first swing instability associated with a single mode; it could be a result of superposition of a slow inter-area swing mode and a local plant swing mode causing a large excursion of rotor angle beyond the first swing [4]. Main factors affecting transient stability include: Generator real and reactive power loadings prior to the fault: these determine the internal voltage magnitude of the generator. Generator output during the fault: depends on the internal voltage magnitude of the generator, and the fault location and type. Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 6

16 Section 2 Transient Stability Criteria and Performance Requirements Fault-clearing time: - depends upon the type of fault, the circuit-breaker operating times and the relay schemes used in the protection operation. Post-fault transmission system reactance: in general, the reactance seen by the least stable generator. The generator design: a low internal reactance increased peak power and reduced initial rotor angle. Generator inertia: for a specific machine, the higher the inertia, the lower the rate of change in angle. This is usually beneficial to overall system stability but it depends on the network and the loading conditions. Transmission loading: A power system with a high-power transfer from one area to another can be more vulnerable to transient stability problems. There are two effects. Firstly, the ability to keep generators synchronised may be reduced by the higher angular separation caused by the power transfer. Secondly, faults between load and generation groups can result in the load part of the network slowing down during a fault while the generator part of the network accelerates; this can increase angular separation and make instability more likely. For more information in relation to transient stability theory, refer to Appendix A System faults For analysing transient stability, 3-phase-to-ground (3PG) zero impedance faults were applied with a post-fault circuit outage. If the 3-phase-to-ground faults applied had a significant impact on transient stability, then 2-phase-to-ground faults were applied for comparison purposes. System faults which have a greater impact on the transient response of the system were investigated. Table 1 and Table 2 show the system faults applied at each 220 kv bus in North and South Islands. Figure 1 shows the geographical representation of the 220 kv bus faults in North Island as summarised in Table 1. Likewise, Figure 2 represents the geographical representation of the 220 kv faults in South Island as summarised in Table 2. Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 7

17 Section 2 Transient Stability Criteria and Performance Requirements Grid exist point Bus volt Fault type Fault duration Bunnythorpe 220 kv 3-phase-to-ground 120 ms Hamilton 220 kv 3-phase-to-ground 120 ms Haywards 220 kv 3-phase-to-ground 120 ms Huntly 220 kv 3-phase-to-ground 120 ms Otahuhu 220 kv 3-phase-to-ground 120 ms Pakuranga 220 kv 3-phase-to-ground 120 ms Stratford 220 kv 3-phase-to-ground 120 ms Whakamaru 220 kv 3-phase-to-ground 120 ms Wairakei 220 kv 3-phase-to-ground 120 ms Table 1: North Island bus faults Grid exist point Bus Volt Fault type Fault duration Ashburton 220 kv 3-phase-to-ground 120 ms Benmore 220 kv 3-phase-to-ground 120 ms Bromley 220 kv 3-phase-to-ground 120 ms Cromwell 220 kv 3-phase-to-ground 120 ms Culverden 220 kv 3-phase-to-ground 120 ms Clyde 220 kv 3-phase-to-ground 120 ms Invercargill 220 kv 3-phase-to-ground 120 ms Islington 220 kv 3-phase-to-ground 120 ms Kikiwa 220 kv 3-phase-to-ground 120 ms Livingstone 220 kv 3-phase-to-ground 120 ms Manapouri 220 kv 3-phase-to-ground 120 ms North Makarewa 220 kv 3-phase-to-ground 120 ms Naseby 220 kv 3-phase-to-ground 120 ms Roxburgh 220 kv 3-phase-to-ground 120 ms Twizel 220 kv 3-phase-to-ground 120 ms Waipara 220 kv 3-phase-to-ground 120 ms Table 2: South Island bus faults Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 8

18 Section 2 Transient Stability Criteria and Performance Requirements Figure 1: 220 kv bus faults simulated in North Island Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 9

19 Section 2 Transient Stability Criteria and Performance Requirements Figure 2: 220 kv bus faults simulated in South Island Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 10

20 Section 2 Transient Stability Criteria and Performance Requirements Fault clearance times Transpower implements the following protection clearing times as a general guideline for system protection. Protection standard at 220 kv and above: The target fault clearance time from main protection is 120 ms. For buses, backup for single main bus protection scheme shall be provided by distance or overcurrent protection at the far end of the transmission circuit or transformer connecting to the bus. Transmission circuits 220 kv and above shall have duplicate main protection with signal co-operation. Protection standard from 50 kv to 100 kv: The target fault clearance time from main protection is 120 ms. However, some parts of the sub-transmission system may have the same degree of importance as the 220 kv system. Where this is determined, Transpower applies the same level of protection to these assets. Historically 200 ms was used as the target, but with modern circuit breakers and protection, 120 ms is used. Protection standard from 11 kv to 33 kv: The target fault clearance time from main protection is 200 ms. For this transient stability assessment, 120 ms was used as the standard protection clearance time for 110 kv and 220 kv transmission circuits and buses. Slow protection operation due to main protection failure, as well as high speed circuit breaker re-closure, are considered beyond the scope of work of this study. 2.3 Criteria used in New Zealand Code requirement The Code does not specify a definite standard or requirements for assessing transient stability of the New Zealand power system. In regards to stability events, the system operator Policy Statement defines stability events as: Severe power system faults that might lead to a defined contingent event, extended contingent event or loss of an interconnecting transformer or bus section. For these faults, it is deemed prudent to ensure that the transient and dynamic stability of the power system is maintained [5] Operational practice for ensuring stable operation As system operator, Transpower manages transient stability by applying permanent operational constraints in the market dispatch system. This involves limiting transfers between regions to a pre-determined value to ensure transient instability does not occur for a fault over all typically encountered power system conditions. Such limits are Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 11

21 Section 2 Transient Stability Criteria and Performance Requirements presently determined in advance, as the calculation is not available in the real-time environment. Two circuits have been identified with transient stability issues: Manapouri and Huntly Stratford. Manapouri At a Manapouri generation intertrip is utilised to ensure system stability. The scheme will trip one Manapouri unit if station generation is above 760 MW and two units if station generation is above 795MW for the simultaneous loss of any two Manapouri 220 kv circuits, or any one Manapouri circuit, where another Manapouri circuit is out of service. This scheme allows unconstrained generation at Manapouri up to its maximum station output in steady-state and enables the transient stability limit at Manapouri to be managed. Equation 1 shows the build of the permanent constraint used to manage transient stability where X is the transfer limit to maintain transient stability. The effect of this manual constraint is to manage flows through Manapouri-North Makarewa-1, Manapouri-North Makarewa-2, Manapouri-North Makarewa-3 and Invercargill-Manapouri-2 circuits to manage transient stability when the Manapouri intertrip scheme is enabled or disabled. (The right-hand side value X of the constraint is subject to take a different value based on factors such as status of the Manapouri intertrip scheme: enabled or disabled, etc.) Equation 1 1 * MAN_NMA * MAN_NMA * MAN_NMA * INV_MAN.1 X MW Huntly-Stratford The transient stability permanent manual constraint is used by Transpower to limit precontingency flows from Stratford to Huntly and Stratford to Taumarunui for a single contingency during high Taranaki generation and high north flow. The build of this constraint is shown in Equation 2. As mentioned above, the value of X can vary depending on the system conditions (e.g. transmission outages). Equation 2-1 * HLY_SFD * SFD_TMN1.1 X MW In the Policy Statement, the following clauses provide for managing a situation where demand shedding is used for grid emergencies that may cause transient or dynamic stability issues: Clause 12.5: If, in the system operator s reasonable opinion, a credible event is likely to lead to a stability event, the system operator may rely on demand shedding to maintain the power system within identified transient and/or dynamic stability limits in accordance with Clause 74. Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 12

22 Section 2 Transient Stability Criteria and Performance Requirements Clause 18: For stability events, the system operator plans to ensure that the transient and dynamic stability of the power system is maintained Clause 74: Demand shedding will be as per Table 3 for transient stability. Scenario Prior to 2 Hours Within 2 hours Demand shedding policy Steady state, including steady stage after an event has occurred (transient or dynamic stability is about to occur). Issue a warning notice. Declare a grid emergency. Demand shedding will occur if participant responses to the generator do not mitigate the grid emergency. For a defined event (transient or dynamic stability limit is being exceeded). Issue a warning notice. Declare a grid emergency. Demand shedding will occur if participant responses to the generator do not mitigate the grid emergency. For a second defined event (after an event has occurred)- transient or dynamic stability limits is being exceeded for a second defined event. Declare a grid emergency. Table 3: Demand shedding to maintain transient stability of New Zealand Demand shedding may occur where the system operator reasonably believes there is a significantly elevated risk of a second defined event. Transpower runs TSAT (Transient Security Assessment Tool) Online to analyse system frequency behaviour for contingencies during real-time operation. At the time of writing this report, TSAT Online has not been setup to carry out real-time transient stability analysis Criteria for assessing system transient performance For all synchronous machines, the stability of the generators is determined by examining their rotor angles. Fundamentally, for stability, the angle between each pair of machines reaches a peak value during the transient and then decreases. If any of these angle differences increases indefinitely, the system is by definition unstable, since at least one machine loses synchronism with the remainder of the system. Typically for a multi-machine power system, the rotor angles are measured against their local bus voltages. Therefore, stability is analysed as a local phenomenon in conventional methods during transient conditions. Generally, in a power system, voltage phase angles vary (if transformer phase shifts are neglected) across the network, and during grid Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 13

23 Section 2 Transient Stability Criteria and Performance Requirements disturbances voltage phase angle distributions are highly variable due to the transient currents in the network. Therefore, in some instances, rotor angle measurements taken in a local frame of reference can lead to an incorrect assessment of transient stability. The rotor angle can be specified with respect to several reference frames such as: Rotor angle w.r.t local bus voltage. Rotor angle w.r.t slack-bus (reference machine) voltage angle. Rotor angle w.r.t slack-bus (reference machine) rotor angle Transient instability Transient instability or rotor angle instability is commonly observed as a failure to recover from the first power swing. When the system loses stability during the first swing, the response of the unstable generator(s) is expected to be as shown in Figure 3. The rotor angle of one or more synchronous machines (relative to other machines) continues to increase until synchronism is lost. Figure 3: Rotor angle response with first swing instability In some countries, criteria are applied to limit the acceptable magnitude of the first rotor angle swing. These criteria are intended to provide a margin to any possible unstable operation but also provide a measure which can be easily incorporated into system studies to produce consistent results Oscillatory instability If the machine is first swing stable, the machine rotor angles will oscillate following the first swing before settling at the stable operating point. If the oscillations do not settle at all or take an excessive amount of time to settle, they are said to have insufficient damping and the system then has oscillatory stability problems. Figure 4 shows oscillatory instability compared with the stable response. Two criteria, the Time constant (T) and the damping ratio (Ϛ) may be used to assess oscillatory stability. These are described below. Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 14

24 Section 2 Transient Stability Criteria and Performance Requirements Figure 4: Rotor angle response with oscillatory instability Time constant (T) is the time for the maximum displacement to decay to 1 of its initial ee value. For this study, a criterion has been adopted based on United Kingdom practice that states the system is considered oscillatory stable if oscillation decay has at most a 12 second time constant [6]. The advantage of using the time constant criteria is that it is conveniently identifiable and can be directly measured from simulations. It also recognises that higher frequency oscillations are of less concern and can therefore be allowed to be more poorly damped. Damping ratio (Ϛ) is the damping of an oscillation and is a measure of how sustained an oscillation will be. The damping ratio is a comparative measure of damping often used when a range of frequencies of oscillation can occur. The damping ratio of the observable system oscillations can be used, in both small signal and transient studies, as an indicative measure of whether the overall system stability is improving or getting worse. Other ways of damping as alternative ways of specifying damping exists as follows [7]: Decay time: defined as 4 x time constant. For example, the amplitude of a transient signal decays to less than 2% of its initial value within four time constants. Number of cycles before settling: Time constant and decay time criteria do not apply the same damping criteria for all frequencies of oscillation. A measure related to the damping ratio which does take account of frequency is the ratio between the decay time and period of oscillation, i.e., the number of cycles of oscillations allowed before a specified magnitude drop has been observed. Decay ratio: The ratio of the magnitude of subsequent peaks in the impulse response. Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 15

25 Section 3 Assumptions and study methodology 3 ASSUMPTIONS AND STUDY METHODOLOGY 3.1 Study assumptions The following assumptions were made for this study: Modelling using a conventional transient (RMS) stability programme was considered sufficient for this study. While most PV installations are single phase, it is the stability of the other (3 phase connected) generation that is of most concern, and the modelling of PV generation can be abstracted to a sequence network representation for this purpose. Transient stability studies are based on all circuits in-service followed by the post-fault disconnection of a transmission circuit. The fault locations and fault types studies have been identified; these are expected to be limited to events which the system operator defines as Credible Events. Energy storage (e.g., a battery installation at the PV site) is not considered, therefore power output from the PV systems has been assessed based on solar variation during the day. PV inverters connected to the system comply with AS/NZS :2015 standard. The ability of inverters to remain connected to the grid during disturbances assumed to meet the requirements in this standard. Corrective actions from current automatic under frequency load shedding schemes have not operated. This is an existing requirement for grid operation. No industrial level solar PV installations such as solar farms/parks are included in the study. Significant motor loads that may have an impact on transient stability or system behaviour under study have not been modelled using dynamic load models. This is considered beyond the scope of this study but is highlighted as an area that requires further consideration. Current capability and configuration of the New Zealand power system has been considered, including committed reinforcements. Non-committed network changes have been ignored. No new generating plants are commissioned beyond those currently committed and no plant closures have been included. Solar PV inverters have 0 Mvar steady-state output. Solar PV inverter behaviour is as described in subsection Solar PV inverters post-fault real power response matches the performance observed during inverter testing carried out at the University of Auckland; specifically, speed of recovery back to the pre-fault power output with no power overshoot are important assumptions. Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 16

26 Section 3 Assumptions and study methodology It is assumed the models provided to Transpower by the respective asset owners have been fully validated for excitation system model, governor model and generator model, and closely represent the actual operation in the field. Ensuring existing models are fully validated and tested is outside the scope of this project. 3.2 Study inputs Power-flow preparation A range of power-flow cases were prepared to investigate impact of PV generation on transient stability of the New Zealand power system for winter and summer seasons. These power-flow cases were generated from the following sunny days (red points in the PV generation curves represent scenarios selected for each study): Winter Tuesday 28/07/2015 Winter Saturday 01/08/2015 Summer Sunday 10/01/2016 Figure 5, Figure 6 and Figure 7 show non-pv (conventional) generation, total system demand, net system demand and PV generation for a typical day with PV generation in winter and summer. These scenarios were developed during the work on the first report produced for the PV Generation Investigation Project - Effect of Solar PV on Generation Dispatch in New Zealand. Based on installed capacity of 4GW total generation, PV generation reaches 3000 MW in winter and 3500 MW in summer. As shown in the figures, PV generation starts to ramp up in the morning from about 0700 hrs, peaking at midday between 1200 hrs and 1300 hrs, then ramping back down as sun set approaches. Some literature refers to the net load curve as a duck curve where high PV penetration has displaced major conventional generation, thereby creating a duck shape load profile. Figure 5: Winter Tuesday 27/07/2015 sunny day scenarios Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 17

27 Section 3 Assumptions and study methodology Figure 6: Winter Saturday 01/08/2015 sunny day scenarios Figure 7: Summer Sunday 10/01/2016 sunny day scenarios Six types of PV inverters were modelled in this study. These types will be explained in more detail in subsection and Appendix A2. Table 4 depicts inverter types and their control modes modelled in the power-flow cases. Inverter Type Under-Frequency Response Over-Frequency Response Reactive Power Control Type 1- Inverter A No response Frequency Control Volt-Var Mode Type 2- Inverter A No response Frequency Control Constant Power Factor Type 3- Inverter B No response Frequency Control Constant Var Type 4- Inverter B No response Constant Power Volt-Var Mode Type 5- Inverter C No response Constant Power Constant Power Factor Type 6- Inverter C No response Constant Power Constant Var Table 4: PV Inverter types Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 18

28 Section 3 Assumptions and study methodology Note: Because the PV inverters are generating at maximum MW output, underfrequency response is not expected from the inverter dynamic models (more details will be discussed in subsection 3.2.3). The proportion of PV inverter types distributed in the New Zealand power system is based on the market share of the tested inverters in the United States [8]. The three tested inverter types accounted for 41% of the total market share in Consequently, extrapolation of their relative market shares to 100%, constitute 17% of inverter A, 71% of inverter B and 12% of inverter C distribution factors. Furthermore, these three inverter market shares were subdivided into reactive power control settings: volt-var mode and power factor control mode, and applied to the national PV distribution in the power-flow scenarios. Inverter Type Inverter Count Percentage Proportion Type 1- Inverter A % Type 2- Inverter A % Type 3- Inverter B % Type 4- Inverter B % Type 5- Inverter C % Type 6- Inverter C % Total % Table 5: PV inverter type allocations Table 5 shows how much of each type of inverter technology (type A, B and C) are used in the power-flows. Inverter count represents the number aggregated PV generating units used in the power-flow. Approximately 70% of PV inverters modelled are of type B. This implies that the dynamic response of Inverter B could have a more significant impact on the system dynamic response than other types. Figure 8 shows the percentage distribution of different inverter types (as mentioned in Table 5) at each Grid Zone in both islands. For example, in the region of Grid Zone 1 the distribution of inverter types is: type %, type 2-7.9%, type %, type %, type 5-5.3% and type 6-7.9%. The figure shows that type 3 and type 4 have the highest percentage number of inverters within a Grid Zone compared to the other inverter types 1, 2, 5 and 6. Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 19

29 Section 3 Assumptions and study methodology Figure 8: Installed PV inverter types as a percentage of total Grid Zone PV inverters Figure 9 shows a typical PV generation pattern for winter Tuesday power-flow cases. This pattern is similar for the winter Saturday and summer Sunday power-flow cases used for the transient stability simulations. The North Island PV generation pattern shows that the majority of generation is concentrated in Auckland region (Grid Zone 2). Northland (Grid Zone 1) shows as having the second highest PV penetration levels followed by Wellington (Grid Zone 8). In the South Island, Grid Zone 10 was expected to have the highest PV penetration followed by Southland (Grid Zone 14) and Nelson (Grid Zone 9). Figure 9 shows the typical PV generation pattern for winter Tuesday power-flow cases. This pattern is similar for the winter Saturday and summer Sunday power-flow cases used for the transient stability simulations. Figure 9: Grid Zone PV penetration levels (winter, Tuesday) In the power-flow, PV inverters were modelled such that an inverter was connected to an 11 kv PV generator bus via a step-up transformer (see Figure 10). The number of aggregated inverters connected to grid buses at grid exit points ranged between 1 to 8, Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 20

30 Section 3 Assumptions and study methodology which was determined by the maximum PV generation at that node in summer and winter scenarios where PV generation was at its peak. Python script was written to generate power-flow cases (*.raw format) with optimised PV inverter MW outputs as shown in Figure 11. Assignment of varying MW output for inverters connected at a particular grid exit point was implemented as a more realistic approach, otherwise having equal MW output for each inverter at the grid exit point (GXP) would have been overly conservative for this study. These power-flow cases were then converted to PSAT (*.pfb format) to be used in TSAT for transient stability simulation studies. As a rule of thumb, the MVA rating of PV inverters was set to 125% of their MW output. Figure 10: Example of a PV inverter grid connection configuration in the power-flow Figure 11: Process used to allocate MW output for each inverter at a given grid exit point Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 21

31 Section 3 Assumptions and study methodology Power-flow scenarios Power-flow scenarios used for simulations are summarised in Table 6. As shown in the table, 23 power-flow cases were prepared with varying North Island installed PV generation, South Island installed PV generation, system demand and HVDC flow. Only three power-flow cases were prepared for the summer study (case 21, case 22 and case 23). This is because to solve the power-flow cases after 1100 hrs with a secure network solution required network modifications (e.g. adding reactors and removing 220 kv transmission circuits) to manage high transmission voltages and reactive power-flows on the transmission circuits. These issues have been highlighted in the related report regarding voltage management and control [9]. The days selected for the scenarios present the most distinctively different daily patterns observed on the power system. The winter Tuesday category represents a typical business working day in winter. Likewise, the winter Saturday and summer Sunday categories represent typical weekend days. Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 22

32 Section 3 Assumptions and study methodology Case ID Case Details Load (MW, MVAR) PV Generation (MW) North Island South Island North Island South Island HVDC (MW) Conventional Units Status North Island South Island Case 1 Winter, Tue 0700 (3508, 561) (1700, 332) North Case 2 Winter, Tue 0800 (3920, 616) (2020, 379) North Case 3 Winter, Tue 1030 (3525, 593) (1677, 459) South Case 4 Winter, Tue 1100 (3482, 587) (1738, 476) South Case 5 Winter, Tue 1200 (3434, 581) (1708, 467) South Case 6 Winter, Tue 1300 (3367, 568) (1691, 463) South Case 7 Winter, Tue 1400 (3324, 561) (1690, 458) South Case 8 Winter, Tue 1500 (3347, 567) (1671, 455) South Case 9 Winter, Tue 1600 (3458, 561) (1749, 352) North Case 10 Winter, Tue 1700 (3742, 589) (1800, 345) North Case 11 Winter, Tue 1800 (4208, 641) (1900, 350) North Case 12 Winter, Sat 0830 (3380, 584) (1700, 465) North Case 13 Winter, Sat 0930 (3541, 612) (1828, 500) South Case 14 Winter, Sat 1030 (3434, 593) (1624, 444) South Case 15 Winter, Sat 1130 (3305, 571) (1672, 459) South Case 16 Winter, Sat 1230 (3246, 562) (1687, 461) South Case 17 Winter, Sat 1330 (3136, 543) (1653, 453) South Refer to Appendix A1: Figure A1-1 Refer to Appendix A1: Refer to Appendix A1: Figure A1-2 Refer to Appendix A1: Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 23

33 Section 3 Assumptions and study methodology Case ID Case Details Load (MW, MVAR) PV Generation (MW) North Island South Island North Island South Island HVDC (MW) Conventional Units Status North Island South Island Case 18 Winter, Sat 1430 (3049, 529) (1608, 441) South Figure A1-1 Figure A1-2 Case 19 Winter, Sat 1530 (3060, 532) (1606,439) South Case 20 Winter, Sat 1600 (3122,544) (1614,441) South Case 21 Case 22 Case 23 Summer, Sun 0800 Summer, Sun 1000 Summer, Sun 1100 (2446,499) (1561,453) South (2825,565) (1684, 483) South (2787, 555) (1675,481) South Table 6: Power-flow scenarios used for the transient stability study Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 24

34 Section 3 Assumptions and study methodology To generate the base cases mentioned above, information was used from the first report produced for the PV Generation Investigation Project - Effect of Solar PV on Generation Dispatch in New Zealand. A total of 71 National Institute of Water and Atmospheric Research (NIWA) sites with reliable irradiance data were used to get an indication of PV generation levels of the 16 representative council regions. These regional PV generation levels were further disaggregated to GXP levels lying within each council region s geographic boundaries. Detailed information of the derivation of solar PV injection levels from solar irradiance information at each GXP is discussed in detail in Transpower s Effect of Solar PV on Generator Dispatch in New Zealand report PV inverter dynamic models Transpower engaged Auckland University to test 3 main types of inverters to better understand the capabilities of the expected behaviour of different inverter technology types: inverter A, inverter B and inverter C. These inverters were specifically tested to identify the following: Active power regulation due to over-frequency and under-frequency. Reactive power regulation (voltage regulation). Voltage and frequency ride-through characteristics. Note: Ride-through is a term used for the ability of equipment to remain connected and return to normal operation following a disturbance. For PV generation, ridethrough means that power output is restored immediately post event and the equipment does not go through a disconnect/re-connect cycle. In addition to specific active power, reactive power and ride-through characteristic tests, the dynamic behaviour of the inverters, specifically how the power and reactive power output varied, was investigated in detail for a range of tests in each of the following areas: Ride-through characteristics Fault response Reconnection characteristics Voltage or reactive power response Frequency response Some of these aspects of the inverter behaviour are described in more detail in subsections and 0 (see also Appendix A4 for full description) Ride-through characteristics Inverter types A, B and C exhibited different characteristics when tested for voltage and frequency ride-through. Inverter type B is most resilient to under-voltage, with tests showing it remaining connected at 10 V (0.043 pu) for 1 second. Inverter type C had an Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 25

35 Section 3 Assumptions and study methodology instantaneous under-voltage trip at 160 V (0.696 pu), and inverter type A had a trip setting at 45V (0.195 pu) for 10 ms. All inverter types are capable of riding through most frequency events expected in the power system. However, inverter types B and C may trip in a South Island over-frequency event, assisting in the arrest of over-frequency. Inverter type C may trip for a South Island extended contingent event (ECE) as the frequency could potentially fall to 45 Hz. The effect of any trips for extended contingencies was not considered in this study as the fault scenarios used were limited to contingent events. The voltage and frequency ride-through capabilities of the inverters have been modelled as zero output limiters on the active and reactive power control blocks, rather than disconnections of the model from the simulation. Figure 73- Appendix A2, shows the control block diagram for frequency tripping, voltage tripping and reconnect settings component of the inverter type C, and also the interface to active power maximum (Pmax) and reactive power maximum (Qmax). Inverter A and B have the same structure with settings as per the Table 7. Inverter C has under-voltage trip settings that are not compliant with the AS/NZS :2015 standard. Consequently, two versions of the model were issued for analysis: one with tested settings (non-compliant model), and another one with inverter type C with a voltage drop of 0.78 PU for 1 second (compliant model). For transient stability studies, inverter type C with the compliant model was used. This is an important assumption and is discussed further in the results and conclusions in section 5. Table 7: Inverter voltage and frequency ride trip settings used in the dynamic models Fault response The power recovery of inverters following a voltage transient is critical to the transient stability result. This power recovery takes place during the first post fault power swing of the system. With high PV generation injected into the grid, it is clear that the power recovery of these devices will be very important. A review of available studies suggests inverter control system models often allow additional real power to be transiently delivered to the network. For instance, this is representative of inverters connecting wind turbines, where there is a physical source of additional power that may be available from the kinetic energy of the turbine. However, for a PV panel this is considered unrealistic and the test results support the observation that the PV panel will at best return to its pre-fault power output (more specifically its Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 26

36 Section 3 Assumptions and study methodology pre-fault current output). These considerations resulted in specific changes to the control of the inverter interfaces as detailed below. A large active power overshoot was not allowed from the dynamic models; the only allowance was a power increase (typically 1 to 2 percent) related to changes in terminal voltage conditions. The PV panel is basically a voltage source with little energy stored in capacitance within the inverter. Therefore, it could provide only a very small amount of additional active power after fault clearance. The input to the active power (P) and reactive power (Q) PI control blocks were zeroed out whenever the inverter was blocked (usually for voltages below 0.4 PU) to prevent integrator wind-up during the blocked period. Figure 12 shows the active power fault response of inverter types A, B and C Active Power (pu) Time (s) Inverter B Terminal Voltage Inverter A Inverter B Inverter C Figure 12: active power fault response of inverter type A, B and C According to the assumption made for this study, the inverters are compliant with the AS/NZS :2015 standard, thus they have the capability to remain in operation for voltage variations with the duration specified in the standard. The test carried out for the voltage disturbance signal is shown in Figure 13. The top graph shows the voltage disturbance signal input to the inverter and the bottom graph shows the RMS current output of the inverter. These ride-through characteristics are assumed for the 3 types of inverters A, B and C used in this work. The voltage step before the disconnection is the minimum low-voltage ride through. Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 27

37 Section 3 Assumptions and study methodology Figure 13: Inverter RMS current output for a voltage disturbance step signal Note: For more information regarding the PV dynamic model responses, refer to Appendix A Study methodology This section describes the study methodology used for assessing transient stability. Analysis of transient stability of power systems involves computation of their non-linear dynamic response to large disturbances, usually a transmission network fault followed by the isolation of the faulted element by protective relaying. Figure 14 shows the general approach used for assessment. With this approach, good electricity industry practice was followed when including other considerations relevant to the transient stability analysis. Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 28

38 Section 3 Assumptions and study methodology Figure 14: High level overview of the transient stability assessment process Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 29

39 Section 3 Assumptions and study methodology Transient study scenario definitions To analyse the PV inverter effect on the system transient stability, several contingent event scenarios were developed, as shown in Table 8. The initial study cases with PV generation at each GIP as well as loads at each GXP were derived during the work on the first report produced for the PV Generation Investigation Project - Effect of Solar PV on Generation Dispatch in New Zealand, as described in subsection Transient stability assessments performed for these study scenarios are discussed in section 4. Island Identifier Study scenario North Island SCN1 3 phase-to-ground fault at Bunnythorpe 220 kv bus and disconnect Bunnythorpe-Tokaanu-1 North Island SCN2 3-phase-to-ground fault at Hamilton 220 kv bus and disconnect Hamilton-Whakamaru-1 North Island SCN3 3-phase-to-ground fault at Haywards 220 bus and disconnect Bunnythorpe-Paraparaumu-Haywards-1 North Island SCN4 3-phase-to-groud fault at Huntly 220 kv bus and remove Huntly-Startford-1 from service. North Island SCN5 3-phase-to-ground fault at Albany 220 kv bus and disconnect Albany-Henderson-1 from service North Island SCN6 3-phase-to-ground fault at Pakuranga 220 kv bus and disconnect Pakuranga-Whakamaru-1 North Island SCN7 3-phase-to-ground fault at Otahuhu 220 bus and disconnect Otahuhu-Whakamaru-1 North Island SCN8 3-phase-to-ground fault at Whakamaru 220 kv bus and disconnect Tokaanu-Whakamaru-1 North Island SCN9 3-phase-to-ground fault at Wairakei 220 kv bus and disconnect Whakamaru-Wairakei-1 North Island SCN10 3-phase-to-ground fault at Startford 220 kv bus and disconnect Huntly-Stratford-1 South Island SCN11 3-phase-to-ground fault at Ashburton 220 kv bus and disconnect Ashburton-Twizel-1 South Island SCN12 3-phase-to-ground fault at Bromley 220 kv bus and disconnect Bromley-Islington-1 South Island SCN13 3-phase-to-ground fault at Cromwell 220 kv bus and disconnect Clyde-Twizel-1 South Island SCN14 3-phase-to-ground fault at Clyde 220 kv bus and disconnect Clyde-Twizel-1 from service Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 30

40 Section 3 Assumptions and study methodology Island Identifier Study scenario South Island SCN15 3-phase-to-ground fault at Culverden 220 kv bus and disconnect Islington-Kikiwa-2 South Island SCN16 3-phase-to-ground fault at Islington 220 kv bus and disconnect Islington-Kikiwa-1 South Island SCN17 3-phase-to-ground fault at Waipara 220 kv bus and disconnect Islington-Kikiwa-3 South Island SCN18 3-phase-to-ground fault at Invercargill 220 kv bus and disconnect Invercargill-Manapouri-1 South Island SCN19 3-phase-to-ground fault at Manapouri 220 kv bus and disconnect Manapouri-North Makarewa-1 and 2 South Island SCN20 3-phase-to-ground fault at North Makarewa 220 kv bus and disconnect Ivercargill-Manapouri-1 South Island SCN21 3-phase-to-ground fault at Livingstone 220 kv bus and disconnect Islington-Livingston-1 South Island SCN22 3-phase-to-ground fault at Naseby 220 kv bus and disconnect Naseby-Roxburgh-1 South Island SCN23 3-phase-to-ground fault at Roxburgh 220 kv bus and disconnect Naseby-Roxburgh-1 South Island SCN24 3-phase-to-ground fault at Twizel 220 kv bus and disconnect Clyde-Twizel-1 Table 8: Transient stability study scenarios Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 31

41 Section 4 Effect on transient stability 4 EFFECT ON TRANSIENT STABILITY To perform system dynamic studies, a 3 phase-to-ground fault was applied at a specific 220 kv bus, as mentioned in Table 1 and Table 2. This fault was then cleared in 120 ms followed by removal of a 220 kv transmission circuit from service. Time-domain monitoring was performed on: Generator angles Generator speeds 220 kv bus voltages 220 kv circuit active and reactive power-flows Bus frequency PV interface active and reactive power These dynamic studies were carried out for typical winter Tuesday, winter Saturday and summer Sunday cases (case 1 to case 23), as listed in Table 6. As explained in subsection , first swing rotor angle or transient instability was the primary focus of this study. However, transient stability indexes were also calculated for each study scenario mentioned in Table 8. This index provided a comparative measure of rotor angle separation between synchronous generators in the network following a transient grid fault. Transient stability index and its associated definitions are described in more detail under subsection 4.2. This extra measure should provide a reasonable insight into the impact of PV generation on the transient stability of the power system; in particular with the reduction in the number of connected conventional generators, reduction in inertia could lead to initial oscillations being less well damped. Not all the 24 study scenarios analysed for the transient stability studies are presented in detail. Two indicative scenarios from each island have been discussed in detail in subsections 4.2, 4.3, 4.4 and 4.5, and the remainder of the scenarios have been consolidated and discussed together in subsections 4.6 and 4.7. When a fault occurs in a large power system, usually only a few machines show a large response to the fault and would be at risk of losing synchronism. These machines are known as the critical machines. Therefore, when doing a wide-ranging set of studies such as these, it is often sufficient to study the behaviour of the critical machines with respect to the remainder of the power system to evaluate the transient stability of the system for a specific fault. However, for two reasons this approach was of limited use for assessing the transient stability of the New Zealand power system with increasing levels of PV injections: 1. The critical machines for stability are often larger gas fired plants and PV generation often displaces this generation. Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 32

42 Section 4 Effect on transient stability 2. PV generation is being connected primarily at the load centres, which reduces power across the grid this can change where the worst-case faults would occur. While reduced power system inertia levels due to displacement of synchronous generation could lower some existing transient stability limits, this is not always observed. These factors varied across all the cases used for the simulation, so the approach of looking at one or two specific generators across the range of studies was not appropriate. 4.1 Transmission line loading To understand transient stability, it is crucial to analyse the effect of PV generation on transmission line loadings. This is because when transmission circuits are heavily loaded, the angle separation between synchronous machines on the sending and receiving end increases, whereas when transmission lines are lightly loaded angular separation decreases. This phenomenon can be further explained by the power-angle equation Equation 5. The equation suggests power transfer increases when the angle δ between the sending and receiving end increases. This is because an increasing δ value, for example between 0 0 and 90 0, produces an increase in the sin(δ). In theory, maximum power transfer can be observed when δ = Therefore, when the circuit loadings are low, angle separation δ between synchronous machines is lower. This means when a fault at a particular location with a particular magnitude is applied, the ability of the synchronous rotors to regain synchronism improves compared to a case where transmission circuits and machines are severely loaded. When the PV active power injection increased during the day, the loadings on the transmission circuits reduced resulting in heavy reactive power-flows between parallel circuits. This was because lightly loaded long transmission lines behave like capacitors, producing reactive power. Figure 15 shows the percentage MVA loading of all high voltage transmission lines in the upper North Island region for winter. These values have been obtained from cases 1 to 20 (see Table 6). It is observed that Kaikohe-Maungatapere 110 kv circuit has higher loading when PV generation increases, which is due to the power-flow being reversed south due to high penetration levels of PV injection in the Northland area. Figure 16 shows percentage loading of all high voltage circuits in the upper South Island region for winter cases 1 to 20. For the transient stability study cases, the South Island was modelled based on approximately 22% of the 4 GW of installed solar PV capacity. The upper South Island showed a decline in circuit loadings as in the upper North Island. The charging effect due to lightly loaded circuits was also as described for upper North Island. Similarly, Figure 17 and Figure 18 show the percentage MVA loadings on transmission lines for the upper North Island and upper South Island. Although the MW Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 33

43 Section 4 Effect on transient stability loading on the transmission circuits are not shown in the figures, a larger reduction in the MW flow was observed in comparison to the MVA loading on the circuits. Figure 15: Loadings on the transmission lines in upper North Island region in winter base cases Figure 16: Loadings on the transmission lines in upper South Island region in Winter base cases Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 34

44 Section 4 Effect on transient stability Figure 17: Loadings on transmission circuits in upper North Island region in Summer base cases Figure 18: Loadings on transmission circuit in upper South Island region in Summer base cases Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 35

45 Section 4 Effect on transient stability 4.2 Three phase-to-ground fault at Wairakei 220 kv bus followed by disconnection of Whakamaru-Wairakei-1 circuit from service In this section, winter Tuesday, winter Saturday and summer Sunday scenarios are discussed for a 3-phase-to-ground 220 kv bus fault applied at the Wairakei 220 kv bus, which was then cleared in 120 ms followed by the disconnection of Whakamaru- Wairakei-1 transmission circuit. Figure 19 shows a schematic of the faulted bus and the contingent circuit considered for this study. The transient stability study scenario id used for this section is SCN9 as summarised in Table 8. The study covered the winter Tuesday scenario cases 1 to 11, winter Saturday scenario cases 12 to 20 and summer Sunday scenario cases 21 to 23, and monitored: generator rotor angle, 220 kv bus voltage, machine speed, system frequency, machine active power, machine reactive power, inverter active power interface and inverter reactive power interface. Figure 19: Portion of the 22kV network in Bay of Plenty region For angular stability, the generators close to the fault have the most significant response. For assessing transient stability of the system, electrically close generators to the fault at Whakamaru 220 kv bus were observed closely, whereas the rest of the synchronous machines were monitored to ensure their responses were acceptable. The greatest challenge of this study was when moving from one case to another, because the generation profile, load profile and PV active power distribution profile for each case varied (see Table 10 and Table 12 in Appendix A1 for the case-setup details for North Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 36

46 Section 4 Effect on transient stability Island). For this study, generators in the Bay of Plenty region that were electrically close to the fault and remained in service in the base cases through a season (e.g., winter Tuesday cases) were selected as critical generators for analysis and comment. Figure 20, Figure 21 and Figure 22 show relative rotor angles of selected generators close to the fault in Bay of Plenty area for winter Tuesday, winter Saturday and summer Sunday scenarios respectively. Rotor angles of the relevant monitored generators are as follows: Winter Tuesday: Kawerau G1, Matahina G1, Mokai G1, Te Mihi G1 and Wairakei G1 Winter Saturday: Kawerau G1, Matahina G1, Mokai G1, Te Mihi G1 and Wairakei G1 Summer Sunday: Kawerau G1, Matahina G1, Te Mihi G1 and Wairakei G1 The first observation from the figures mentioned above is that rotor angles of the subject generators did not show first swing instability. During the fault, rotor angles deviated from the remote generators but as the fault was cleared, the generators retained synchronism and the rotor angle oscillations were adequately damped. Figure 74, Figure 75 and Figure 76 in Appendix A3 show rotor angle response of the remaining North Island generators for this study scenario. It shows that in-service generators demonstrated transiently stable operation. Another important observation from the simulation is that in the winter Tuesday scenario (Figure 20), some generators in high PV generation cases showed better transient response than some cases with low PV generation (Wairakei G1 relative rotor angle excursions are smaller in high PV generation cases). Wairakei G1 generator is located very close to the faulted Wairakei 220 kv bus. For the low PV generation cases, the Wairakei generators electrically close to the fault operated at high active power output, resulting in a higher loading in the transmission circuits connected to the generating bus. This created a higher angular separation between this group of generators and the remainder of the North Island generators. By comparison, the high PV generation cases were found to cause the Wairakei generators to operate at a lower active power output displaced by PV generation closer to the loads. The transmission circuits were lightly loaded having less active power transferred from the generator into the grid. In addition, the fast post-recovery of PV inverter active power provided significant restoration of the power system to a new equilibrium point. All these factors may have had the effect of reducing the angular separation and consequently slightly improving the transient performance of the power system. Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 37

47 Section 4 Effect on transient stability Figure 20: Rotor angle of Bay Wairakei G1 generating unit. Ref gen. Glenbrook G1 (winter Tuesday) Figure 21: rotor angle of Matahina G1 generating unit. Ref. gen Glenbrook G1 (winter Saturday) Figure 22: rotor angle of Te Mihi G1 generator. Ref. gen Glenbrook G1 (summer Sunday) Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 38

48 Section 4 Effect on transient stability As explained earlier, when PV generation is increased, the study models the inverters to be compliant for fault ride-through, so most of the power lost during a fault will be quickly recovered post-fault by PV active power response (see Figure 23). While PV active power dropped significantly during the fault in the North Island, it recovered up to 85-90% of the MW value very quickly following fault-clearance (Table 9 shows MW recovered 50 ms after the fault was cleared). This means the rotor angles of the remaining generators would have experienced a minimal deceleration or speed up effect due to the fault. Furthermore, rotor angular separation between generators was lower in high-generation PV cases in comparison to some low-generation PV cases (note there was also some variability in the synchronous generation profile in each case). Fault recovery of the inverters is also non-oscillatory, so on recovery PV generation does not participate in any oscillations of the conventional generation. This is inherent in the technology. If transmission voltage at a conventional generator is subjected to an angle swing, the machine will exhibit a real power swing, because the real power transfer from the mechanical turbine to the electrical system is transmitted through the electromagnetic coupling of the air gap of the generator, and the mechanical system is rotating and has considerable inertia. An inverter-connected voltage source such as a PV panel has no inertia and no electromagnetic coupling. Changes in the transmission voltage angle are tracked by the phase locking control of the inverter and no change in real power output would occur. In effect, PV generation can respond as a conventional generator with a perfect power stabilisation (PSS). In Figure 21 and Figure 22, the relative rotor angle excursions of Matahina G1 and Te Mihi G1 generating units have become slightly larger as PV generation increased. This means the transient performance of the some neighbouring generators could become somewhat weakened when PV generation level increases. Even though the generator output of Matahina and Te Mihi generating units have reduced, the generator behaviour is dependent on the way PV generation has displaced other synchronising generators. This is an example of those generators that exhibit slightly degraded transient behaviour due to increase in PV generation. Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 39

49 Section 4 Effect on transient stability Figure 23: North Island total PV MW output response monitored as an interface Table 9: North Island PV interface MW flow (winter Tuesday) Figure 24 shows the machine speed of the Wairakei G1 generating unit for a number of cases in winter, the unit being very close to the fault applied at Wairakei 220 kv bus. Machine speed increased when the fault was applied, but when the fault was cleared, after a few oscillations the machine speed stabilised. Initial deviations of machine speed were larger in low-pv cases compared with the high-pv cases. This is due to the following: 1. The MW output of Wairakei generating units displaced by PV generation (see Figure 25). 2. The fault recovery of the PV inverters is much faster than conventional generators (see Figure 23- the active power recovered very fast with no oscillations). 3. Since Wairakei generators are very close to the fault, these units will contribute to feeding the fault. A higher mechanical output in pre-fault condition results in a larger mechanical accelerating torque post fault-clearance, causing the rotor angle and generator speed to deviate further. The winter Saturday and summer Sunday cases demonstrated similar frequency behaviour as for the winter Tuesday cases described above. However, generator speed excursions in the two scenarios were slightly larger for high PV generation cases. Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 40

50 Section 4 Effect on transient stability Figure 24: Wairakei G1 machine speed (winter Tuesday) Figure 25: Wairakei station generation (winter Tuesday) To understand the system response in general, power angle-based transient stability indexes were calculated for each system fault scenario. Transient stability index can be defined as follows: η = 360 δδ mmmmmm 360+δδ mmmmmm xx < η < 100 Where δmax is the maximum angle separation of any two generators in the island at the same time, in the post-fault response. The transient stability index for the system was taken as the smallest index amount in all islands. Thus, η > 0 and η 0 correspond to stable and unstable conditions, respectively, which clarifies the definition of the index. The index values closer to 100% can be considered to be more stable, since the angular separations between the synchronous machines in the system are smaller. Using DSAT for the calculation of transient stability index, rotor angles are measured with respect to the local bus voltage angles. Transient stability index (η) is directly proportional to system angle separation (δmax), so providing a good indication of the transient behaviour of the power system following a contingency [10]. Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 41

51 Section 4 Effect on transient stability Figure 26 shows transient stability index for various values of the system angle separation (δmax) Figure 26: Transient stability index for various δmax values To obtain a different perspective of this study scenario, transient stability indexes were calculated for the Wairakei 220kV fault. Figure 27, Figure 28 and Figure 29 show the transient stability indexes calculated for the North Island generators using TSAT for this study scenario. In the winter Tuesday scenario (Figure 27), the transient stability index is higher for the morning and evening cases where PV generation is very low, which means the power system is transiently more stable (e.g., cases 1, 2, 10 and 11). This effect is due to the morning and evening ramp-up in the system demand which forces additional synchronous generators to come online. Likewise, in winter Saturday (Figure 28) and summer Sunday (Figure 29), the overall power system is more stable when PV generation is very low during the morning ramp-up. The stability then slightly reduces (e.g., in case 13 and case 22), and again starts to increase as the PV penetration level increases. Figure 27: Transient stability indexes for the Wairakei 220 kv fault (winter Tuesday) Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 42

52 Section 4 Effect on transient stability Figure 28: Transient stability indexes for the Wairakei 220 kv fault (winter Saturday) Figure 29: Transient stability indexes for Wairakei 220 kv fault (summer Sunday) 4.3 Three phase-to-ground fault at Huntly 220 kv bus followed by disconnection of Huntly-Stratford-1 line from service In this section, winter Tuesday, winter Saturday and summer Sunday scenarios are discussed together for transient study scenario SCN4 (see Table 8). The study covered winter Tuesday scenario cases 1 to 11, winter Saturday scenario case 12 to case 20 and summer Sunday scenario cases 21 to 23. The study applied a 3-phase-to-ground fault close to Huntly 220 kv bus for 120 ms duration (6 cycles). The fault was then cleared followed by disconnection of the Huntly- Stratford-1 transmission circuit. See Figure 30 for the single line diagram of the faulted 220 kv Huntly bus and contingent 220 kv circuit Huntly-Stratford-1. Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 43

53 Section 4 Effect on transient stability Figure 30: Portion of the 220 kv network between Taranaki and Auckland Figure 31, Figure 32 and Figure 33 show relative rotor angles of a selected North Island generators for the 3 phase-to-ground fault at Huntly 220 kv bus, for winter Tuesday, winter Saturday and summer Sunday scenarios respectively. Rotor angles of the following generators were also monitored closely for each study. These generators were selected as being electrically close to the faulted bus and in the study they remained in-service through all the cases for a specific season scenario (e.g. winter Tuesday case 1 to case 11). Winter Tue: Glenbrook G1, Arapuni G1, Mokai G1 and Te Rapa GT1 Winter Sat: Te Rapa GT1, Arapuni G1, Mokai G1 and Wairakei G1 Summer Sun: Glenbrook G1, Arapuni G1, Mokai G1, Te Rapa GT1 and Maraetai G1 Rotor angles of the above generating units started to oscillate but settled to a new stable operating point. None of the generators lost synchronism and first swing instability did not occur. Figure 77, Figure 78 and Figure 79 in Appendix A3 show rotor angles of the remaining in-service synchronous machines for winter Tuesday, winter Saturday and Summer Sunday scenarios, confirming that all the North Island generators retained synchronism. In all three scenarios, rotor angle oscillations of some generators appeared to be more damped for high PV generation cases compared to low PV generation cases. However, this was not always true, as can be seen from Figure 33. It shows that in the summer scenario, when going from case 20, 21 and 23 consecutively, rotor angle oscillations gradually increased. Therefore, it is not possible to generalise that the rotor angle Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 44

54 Section 4 Effect on transient stability response improved with increased PV generation levels. This difference in the behaviour is mainly due to the variations in the conventional generation displacement by PV generation taking place close to the fault. This is discussed next. The magnitude of oscillations can also depend on the generation profile and the way conventional generation was displaced off by PV generation. For example, large generating units such as Huntly UN5, Stratford G21 and G22 etc. were displaced off by PV generation as soon as PV generation started to ramp up in the morning (refer to Table 10 and Table 12 in Appendix A1 for the North Island generation setup for case 1 to case 23). In addition, power-flow on major 220 kv and 110 kv circuits from the lower North Island (on all circuits from Bunnythorpe right up to Auckland) have significantly reduced, thus making them lightly loaded. The general observation regarding transient stability was that as transmission circuits became more lightly loaded, angular separation between inservice conventional generators lessened (see section 4.1) and this is expected to improve the rotor angle stability of some generators. The time constant for each case for this scenario was also calculated and remained within 1 second. Figure 31: Rotor angle Mokai G1 generator. Ref. gen. Glenbrook G1 (winter Tuesday) Figure 32: Rotor angle Wairakei G1 generating unit. Ref. gen Glenbrook G1 (winter Saturday) Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 45

55 Section 4 Effect on transient stability Figure 33: Rotor angle Te Mihi G1 generator. Reg. gen. Glenbrook (summer Sunday) Figure 34 shows Maraetai G2 machine-speed response for the winter Tuesday scenario. Similar response was observed for winter Saturday and summer Sunday scenarios. When the fault was applied, the generator accelerated, as observed by the initial speed increase. If the mechanical power Pmech, electrical power out Pgen and total power loss Ploss, the rotor speed increased when Pmech - Pgen - Ploss > 0 and decreased when Pmech - Pgen - Ploss < 0. The Maraetai generating unit output was displaced when PV generation was increased. This means that initial mechanical power is lower, causing smaller speed variations due to the fault. As seen in the winter Tuesday scenario, when PV generation increased, machine speed excursions reduced. PV generation gradually displaced Maraetei station generation as shown in Figure 35. This result shows smaller excursions in the generator speed for a fault. Figure 36, Figure 37 and Figure 38 show PV inverter power response in North Island monitored as an interface (note that MW values are shown as negative because the total MW output of the inverters was monitored as an interface). As explained in the previous section, PV inverters regained their active power very quickly as soon as the fault was cleared. With conventional generation, electrically close generators tend to oscillate as a group and often it is the magnitude of the real power swings from such groups of generators that is problematic. As previously noted for the first set of studies, PV generation does not participate in any oscillations of the conventional generation. In most of the cases studied, this reduced the size of the machine groups that are oscillating and was beneficial for overall transient stability. Therefore, it is evident that the observed transient stability of the power system generators was as good or better when PV injection was increased compared to the zero or low PV generation cases. Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 46

56 Section 4 Effect on transient stability Figure 34: Maraetai G2 machines speed (winter Tuesday) Figure 35: Maraetai station generation (winter Tuesday) Figure 36: North Island PV inverter MW response (winter Tuesday) Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 47

57 Section 4 Effect on transient stability Figure 37: North Island PV inverter MW response (winter Saturday) Figure 38: North Island PV Inverter MW response (summer Sunday) To gain a better understanding of this study scenario over the entire power system, transient stability indexes were calculated for the North Island generators for the Huntly 220 kv bus fault (see subsection 4.2 for the definition of the transient stability index). This index provides a comparative measure of rotor angle separation between synchronous generators in the North Island for the above fault. Figure 39, Figure 40 and Figure 41 depict the transient stability indexes calculated for the North Island generators using TSAT for the study scenario. In all the scenarios, the transient stability index is higher for the morning and evening cases where PV generation is very low, which means the power system is transiently more stable. This effect is due to the morning and evening ramp-up in system demand which forces additional synchronous generators to come online. During ramp-up periods, system demand is usually high, causing high power transfers on circuits with low amounts of PV generation. Therefore, having more synchronised generators online help improve transient stability (note that in these cases, PV generation is not high enough to significantly reduce the power-flow on the transmission circuits). Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 48

58 Section 4 Effect on transient stability Overall, the power system is more stable when PV generation is very low during the morning and evening ramp-up, after which transient response decreases (e.g., case 3 and case 13), and again starts to increase as PV penetration levels increase. Figure 39: Transient stability indexes for the Huntly 220 kv fault (winter Tuesday) Figure 40: Transient stability indexes for the Huntly 220 kv fault (winter Saturday) Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 49

59 Section 4 Effect on transient stability Figure 41: Transient stability indexes for the Huntly 220 kv fault (summer Sunday) Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 50

60 Section 4 Effect on transient stability 4.4 Three phase-to-ground fault at Clyde 220 kv bus followed by disconnection of Clyde-Twizel-1 circuit from service In this section, winter Tuesday, winter Saturday and summer Sunday scenarios are discussed for a 3-phase-to-ground 220 kv bus fault applied close to the Clyde 220 kv bus which was then cleared in 120 ms followed by removal of the Clyde-Twizel-1 transmission circuit. The transient stability scenario id for this study is SCN14, as summarised in Table 8. The study was performed for the winter Tuesday scenario: cases 1 to 11, winter Saturday scenario: cases 12 to 20 and summer Sunday scenario: cases 21 to 23. Figure 42 shows a schematic of the faulted bus and the contingent circuit. Figure 42: Portion of the 22kV network in Southland and Otago Figure 43, Figure 44 and Figure 45 show simulation results for rotor angles response for a 10 second period of the Clyde G1, Roxburgh G1, Ohau A G4 and Ohau B (G10 and G8) when the 3-phase-to-ground transient fault near the Clyde 220 kv bus was simulated. Relative rotor angle response of Clyde G1 generating unit showed smaller excursions in high-pv-generating cases (Figure 43) compared to low PV generation cases. This is mainly due to the displacement of Clyde generating units and their MW outputs. When PV generation was increased, lower South Island generators were lightly loaded and loadings on Clyde-Twizel-1 and 2 circuits reduced, resulting in the reversal of the power-flow direction (see Figure 47). Generators close to the faulted bus, such as the Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 51

61 Section 4 Effect on transient stability Clyde unit, experienced a lower load in high-pv cases, and with low MW output, the rotor acceleration torque of the generating unit was consequently reduced making it stabilise quickly. Similar behaviour was observed for Roxburgh G1 generating unit, as shown in Figure 44. The relative rotor angle response of the Ohau A -G4 generating unit showed larger oscillations in high PV generation cases. Ohau A is located to the north of the faulted bus. A reverse in power-flow direction between case 3 and case 8 (Figure 47) means Ohau A generating unit was on the sending end of the power. Although power output of Ohau A generating unit is almost similar in high PV generation cases, with the increase in powerflow into south, the rotor angle response became less stable in high PV generation cases. In contrast (see Figure 46), relative rotor angle of some of the high-pv cases exhibited bigger oscillations, whereas others showed lower oscillations. For example, case 7 oscillations were much larger than in case 6. The time constant for each case in this scenario was also calculated and remained within 1 second. As mentioned above, although the initial oscillations during the fault were higher in few cases with high PV generation level, rotor angles reached a new stable operating equilibrium point very quickly after the fault was cleared. All synchronous machines exhibited stable responses and none of the units lost their synchronism. (Figure 80, Figure 81 and Figure 82 in Appendix A3 show the rotor angle response of the remaining generating units in the South Island). As explained in sections 4.2 and 4.3PV inverters responded to a fault by reducing their active power output during the fault but recovered back to pre-fault power almost immediately and without oscillation as soon as the fault was cleared. This fast response of the PV inverters positively impacted some synchronous generators. In this case, during the fault almost no power was transferred by the critical generators very close to the fault such as Clyde G1, as the bus voltages near the fault almost dropped to zero during the faulted period (see Figure 48). During this time, the energy accumulated in the rotor of the Clyde generators causing them to accelerate (note that high generator MW output means more accelerating torque on the rotor than when the MW output is low). However, as a result of PV generation displacing much online synchronous generation, and in conjunction with the reduction of the net system load north and south of the fault location, the machine rotor oscillations both north and south of the fault were stable and well-damped. In addition, reductions in the transmission line loadings may have caused South Island generators to behave like a single group of generators. This is because when the fault was applied, there was no occurrence of a noticeable deceleration in the remote generators which would have otherwise significantly worsened the angular separation of some of the synchronous generators. In the winter Tuesday scenario, rotor angle oscillations were larger for high PV generation cases. In the winter Saturday scenario, rotor angle responses were smaller in high PV Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 52

62 Section 4 Effect on transient stability generation cases. These differences were noticed as explained above, due to the variability in generation mix, PV generation profile, circuit loading and the way the conventional generation was displaced by PV generation. Figure 43: Rotor angle of Clyde G1. Ref. gen. Benmore G1 (winter Tuesday) Figure 44: Rotor angle of Roxburgh G1, Ref. gen. Benmore G1 (winter Tuesday) Figure 45: Rotor angle of Ohau A- G4. Ref. gen. Benmore G1 (winter Tuesday) Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 53

63 Section 4 Effect on transient stability Figure 46: Rotor angle of Ohau B- G10 (winter Tuesday) Figure 47: Clyde-Twizel-1 & 2 MW flow, Clyde station MW and num. of Clyde units online (winter Tuesday) Figure 48: Bus voltage at Clyde 220 kv faulted bus Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 54

64 Section 4 Effect on transient stability Figure 49 shows the machine speed of Clyde G1 generating unit for the winter Tuesday scenario. Machine speed excursion is smaller for high PV generation cases (e.g., case 4, 5, 6 and 7). This corresponds to the smaller rotor angle excursions with high PV generation cases, as discussed above. Responses seen for winter Tuesday and summer Sunday exhibited a similar behaviour and machine speeds regained stable operation quickly. In summary, for the fault investigated in this section, the power system maintained transient stability and PV generation did not have a significant detrimental effect on the transient stability behaviour. Figure 49: Machine speed of Clyde G1 generating unit (winter Tuesday) To get a better understanding of this study scenario from across the entire power system, transient stability indexes were calculated for the South Island generators for the Clyde 220 kv bus fault (see subsection 4.2 for the definition of the transient stability index). This index provides a comparative measure of rotor angle separation between synchronous generators in the South Island for the above fault. Figure 50, Figure 51 and Figure 52 show the transient stability indexes calculated for South Island generators for winter Tuesday, winter Saturday and summer Sunday scenarios respectively. The study showed that when PV generation increases, transient stability index increases, indicating that the largest rotor angle separation between generators in the South Island decreases with increasing levels of PV generation. When PV generation is at its maximum penetration level (e.g., case 5, case 16 and case 23), the overall power system transient performance improves. Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 55

65 Section 4 Effect on transient stability Figure 50: Transient stability indexes for the Clyde 220 kv fault (winter Tuesday) Figure 51: Transient stability indexes for the Clyde 220 kv fault (winter Saturday) Figure 52: Transient stability indexes for the Clyde 220 kv fault (summer Sunday) Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 56

66 Section 4 Effect on transient stability 4.5 Three phase-to-ground fault at Manapouri 220 kv bus followed by disconnection of Manapouri-North Makarewa-1 and 2 circuits This section discusses a three-phase-to-ground fault applied at the Manapouri 220 kv bus followed by disconnection of the Manapouri-North Makarewa-1 and 2 transmission circuits. The fault is applied for 120 ms (6 cycles) by opening line circuit breakers at Manapouri and North Makarewa ends. The Transient study id discussed in this section is SCN19, as mentioned in Table 8. Figure 53 shows the schematic of the faulted bus and contingent transmission circuit situated in lower South Island 220 kv network. Figure 53: Lower South Island 220 kv network A double circuit contingency was considered for this study because this double circuit is often treated as a single contingency by the system operator. This occurs when there is a higher likelihood of double circuit faults occurring based on historical information or on environmental or system conditions. The main trigger for making Manapouri-North Makarewa-1 and 2 circuits double circuit contingent risk is when there is an electrical storm in the near vicinity. Figure 54, Figure 55, Figure 56 and Figure 57 depict rotor angle and machine speed of Manapouri G1 generating units following a fault near the Manapouri 220 kv bus. In most of the low PV generation cases, a higher number of Manapouri generating units remained connected to the grid in the base cases and the power export north was higher. Clyde and Roxburgh stations were dispatched as mentioned in Appendix A1.2 and A1.4. Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 57

67 Section 4 Effect on transient stability For the winter Tuesday scenario, cases with low PV generation levels (e.g., 1, 2, 3, 9, 10 and 11) exhibited much larger rotor angle accelerations (see Figure 54). Accelerations in rotor angle result in a change in generator speed. Consequently, the generators which have large angle differences show larger machine speed excursions from the system speed (see Figure 55). Although cases 3 and 9 are stable they are too close to instability and would not be acceptable operating conditions. These cases indicate that for those operating conditions, the constraint applied in setting up the study case was insufficient. Operationally, the allowed export from Manapouri for these cases would have been stable. Manapouri machine speed accelerated during the fault then started to decelerate after the fault and they are all first swing stable. With the high PV generation cases 5, 6, 7 and 8 in winter Tuesday, the rotor angle deviations are smaller, and so are their machine speed excursions. The main factors which affected the reduction in rotor angle in high PV generation cases are described in the following. In low PV generation cases: Manapouri is exporting more power into the north through contingent circuits; and The MW output of online Manapouri generating units are higher. Therefore, when the Manapouri fault was applied close to the 220 kv bus, mechanical power output (mechanical torque) caused the generating units to accelerate faster. This was due to bus voltage at the Manapouri bus reducing to zero, which meant no power was transferred during the fault. Figure 54: Rotor angle Manapouri G1 (winter Tuesday) Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 58

68 Section 4 Effect on transient stability Figure 55: Machine speed Manapouri G1 (winter Tuesday) In winter Saturday scenarios, Manapouri generating units remained in synchronism. The initial rotor angle excursion was significantly larger (e.g. case 12) in low PV generation cases when compared to the high PV generation cases such as case 15, 16 and 17 (see Figure 56). The initial machine speed acceleration was slightly larger for the low PV generation cases but the difference was small. As with the winter Tuesday scenario, rotor angle and speed excursions were larger in low PV generation cases (see Figure 56 and Figure 57). In contrast, remote generating unit Argyle G1 showed a higher rate of change of frequency in the high PV generation case (see Figure 58) but the remainder of the remote generators showed similar responses, as explained earlier. Observation from the analysis for the summer Sunday scenario showed similar behaviour to the winter Saturday scenario. When PV generation increased, relative rotor angle separation was reduced, as were the machine speed excursions (see Figure 59 and Figure 60). As mentioned earlier, in all three scenarios, reduction in the Manapouri MW output due to MW being displaced by PV generation, caused reduction in power transfer on the Manapouri-North Makarewa and Manapouri-Invercargill circuits, thereby improving rotor angle oscillations of the Manapouri generating units. Though not shown in the figures, the responses of the remaining Manapouri generating units G2, G3, G5, G6 and G7 were identical to G1. Figure 83, Figure 84 and Figure 85 show the rotor angle response of the remainder of the in-service generators in the South Island for this study scenario. Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 59

69 Section 4 Effect on transient stability Figure 56: Rotor angle Manapouri G1 (winter Saturday) Figure 57: Machine speed Manapouri G1 (winter Saturday) Figure 58: Machine speed of Argyle G1 (winter Saturday) Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 60

70 Section 4 Effect on transient stability Figure 59: Rotor angle Manapouri G1 (summer Sunday) Figure 60: Machine speed of Manapouri G1 (summer Sunday) To get a better understanding of this study scenario across the entire power system, transient stability indexes were calculated for the South Island generators for the Manapouri 220 kv bus fault (see subsection 4.2 for the definition of the transient stability index). This index provides a comparative measure of rotor angle separation between synchronous generators in the South Island for the above fault. Figure 61, Figure 62 and Figure 63 show the transient stability indexes calculated for South Island generators using TSAT. The study showed that in all three scenarios, as PV generation increases, the transient stability index increases. This means the power system transient performance improved (smaller angular differences between the generator rotor angles in South Island). The underlying reasons for the improvement of the transient stability in this study scenario is the displacement of Manapouri generating units by increasing PV generation, as explained above. Transient stability indexes in this case confirm that the entire South Island power system becomes more transiently stable with increasing PV injection levels. Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 61

71 Section 4 Effect on transient stability Figure 61: Transient stability indexes for the Manapouri 220 kv fault (winter Tuesday) Figure 62: Transient stability indexes for the Manapouri 220 kv fault (winter Saturday) Figure 63: Transient stability indexes for Manapouri 220 kv fault (summer Sunday) Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 62

72 Section 4 Effect on transient stability 4.6 North Island other scenarios This section discusses the remainder of the fault scenarios in Table 8. For the North Island, the scenarios are SCN1 to SCN3, SCN5 to SCN8 and SCN10. The main reason for simulating a number of fault scenarios was to identify system performance with varying PV injection levels. Compared to a traditional study method, the application of PV generation to the cases, with its inherent stochastic nature, was problematic, particularly where one case became different from another by a number of factors; including, synchronous generation profile, load profile, PV inverter profile and HVDC transfer profile. Having a number of such variables made the transient stability assessments challenging, especially when comparing various cases. To address this issue, it was appropriate to carry out a number of fault studies to determine system behaviour and impacts of various PV penetration levels on transient stability. In addition, analysing generating units using quantities such as rotor angle, machine speed etc., became difficult, as most of the conventional generators were displaced off by PV generation, leaving only a limited number of generators online to be compared with one another across the various cases as PV penetration levels varied. For various faults, the system did not show a noticeable change or detrimental effect on transient stability. However, it was observed that rotor angle excursions of some generators varied across different faults (see Figure 64- which shows rotor angle of Mokai G1 generating unit of case 5 for various faults). Those differences are mainly due to the change in fault locations and fault levels. Faults closer to the generators experienced more severe effects compared to remote generators or to the generators that are electrically far from the fault location. For example, the rotor angle oscillations are larger in SCN8 because the Whakamaru fault is much closer to the Mokai generating unit. Figure 64: Rotor angle of Mokai G1 for various faults (winter Tuesday, case 5) Transient stability indexes for the winter Tuesday with various fault scenarios are shown in the Figure 65. In general, the indexes are higher during the morning and evening system demand ramp-up periods. From case 2 to case 3, the stability index reduces but Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 63

73 Section 4 Effect on transient stability starts to increase again until PV generation is at its maximum (e.g., cases 5 and 6). Thereafter, as PV generation reduces in the afternoon, so does the transient stability index. Overall, generators in the North Island retained transient stability. Figure 65: Transient stability indexes of various fault scenarios in North Island (winter Tuesday) 4.7 South Island - other scenarios This section discusses the other fault scenarios simulated for the South Island. These are SCN11 to SCN13, SCN15 to SCN18 and SCN20 to SCN24. As mentioned in section 4.6, a number of simulations were carried out to get an understanding of the power system behaviour due to the challenges created by introducing PV generation. The variations created in the generation profiles, the net loads on the system and the HVDC transfer, in each case made comparison between cases challenging. None of the generators in the South Island lost synchronism in any of the cases (the system was transiently stable). South Island PV generation is considerably lower than North Island. In the transient stability study cases, the South Island constituted only about 22% of the 4GW of installed solar PV capacity modelled. Therefore, the number of generating units displaced by PV generation was not as great as in the North Island for the same power-flow cases (case 1 to 23). In general, there were less rotor angle excursions of the conventional generators remaining connected in high PV generation cases. For example, Figure 66 shows the rotor angle of Roxburgh G1 generating unit for SCN23 for the winter Tuesday scenario. The rotor angle excursions of high PV generation cases (e.g., cases 5, 6 and 8) are less significant than in low PV generation cases (e.g. cases 1, 2, 10 and 11). Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 64

74 Section 4 Effect on transient stability Figure 66: Rotor angle of Roxburgh G1 unit for SCN23 (winter Tuesday) Transient stability indexes for the winter Saturday with various fault scenarios are shown in Figure 67. In the South Island, when PV generation increased, the transient stability increased for various scenarios. In other words, from the aspect of the entire power system, transient performance of the South Island generators improved when PV generation increased (higher transient stability index means the rotor angular separation between the generators for that case is smaller compared to a case with a lower transient stability index). The displacement of large conventional generating units in the South Island by PV generation and the reduction in the transmission circuit loading make rotor angular separation between in-service synchronous generators smaller. Thus, the South Island power system is beneficially impacted by the increase of PV generation. Figure 67: Transient stability indexes of various fault scenarios in South Island (winter Saturday) Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 65

75 Section 5 Key findings and conclusions 5 KEY FINDINGS AND CONCLUSIONS 5.1 Transient stability of the power system This report records the study done for the transient stability of the New Zealand power system under varying levels of PV penetration. A number of power-flow base cases were prepared and a number of fault simulations were carried out; monitoring rotor angle, machine speed, bus voltages, PV inverter active power interface and PV inverter reactive power interface. Dynamic models for three types of technologies were developed and used in dynamic simulations. The study was intensely focused on investigating whether increased PV generation would have a detrimental effect on the current transiently stable power system. Transient stability is the ability of the power system to remain in synchronism after being subjected to severe disturbances such as faults. Transient stability may be an issue in cases of larger disturbances, such as disconnection of a circuit or a fault where the voltage drops close to zero at the fault location. Dynamic simulations carried out for this project were mainly performed by application of a three phase-to-ground fault at 220 kv bus followed by disconnection of one or two transmission circuits. The results of the simulations and engineering assessments performed on those results showed there is an effect on the transient stability of the New Zealand power system with introduction of non-grid connected solar PV inverters into the system. Solar PV inverters demonstrated certain behaviours during faults, as explained in section Appendix A4. The introduction of higher levels of PV generation caused displacement of existing conventional generators. This occurred in the form of displacing off various synchronous generating units while altering or reducing the output of the remaining online generators. As PV generation was increased, circuit loading on transmission circuits significantly reduced and in some cases flow direction was reversed. A good example seen in the power-flow cases was a reversal in the direction of HVDC flow to south. All these changes in the power-flow caused by increased PV generation affected the dynamic performance of the generating unit and their behaviour during and post fault conditions. The first swing instability did not occur in either the North Island or South Island under any of the scenarios considered in this study. This showed that even with a high penetration level of PV generation into the New Zealand power system, transient instability is unlikely to occur. Overall, in most of the cases studied, generating units which remained online throughout a study scenario (e.g., winter Tuesday) showed an improvement in rotor angle response with high PV penetration levels, compared to that of the cases with low PV penetration levels. In contrast, there were a few generators in which rotor angle separation became somewhat larger for certain fault conditions when PV generation injection was increased. Almost 78% out of 4GW total New Zealand PV generation was presumed based in North Island. Because of this, the dynamic behaviour of the North Island power system was more affected than the South Island power system. Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 66

76 Section 5 Key findings and conclusions Thus, along with the assumptions mentioned in section 3, it is reasonable to infer that with higher penetration levels of PV generation, the transient stability of the New Zealand power system is as good at present or can even show improvement. In other words, high penetration of PV generation is unlikely to cause a detrimental effect on the transient stability of the New Zealand power system. However, it is strongly recommended that more detailed and accurate modelling be conducted in the future when more data and information is available to assess the impact of PV generation on the transient stability limits of the New Zealand power system (see section 6). Finally, the results of the simulations conducted for this transient stability study revealed that PV penetration levels, system topology, type of the disturbance as well as the location of a fault are all important factors in determining the nature of the impact of high PV penetration on the system. 5.2 Effect of low inertia As seen in this study, increasing PV generation levels displace other generation and reduce the amount of conventional rotating generators on the network. System inertia is proportional to the sum of stored kinetic energy in rotating machines that is absorbed or released in response to changes in the power system frequency. Significant increase in PV generation results in a significant reduction in the number of grid-connected synchronous generators in the power system. Therefore, in this study, high PV generation power-flow cases showed lower system inertia than the low PV generation cases, due to a lower number of these rotating masses. In other words, high PV generating cases have lower system inertia than low PV generating cases. In general, during high inertia scenarios (e.g., many synchronous machines online), the initial rate of change of speed deviations caused by faults should be lower than low inertia scenarios (few synchronous machines online). However, the study shows that the increase in distributed PV generation in the New Zealand power system has a beneficial impact on speed excursions due to faults. This is because when PV generation increased, the system behaviour is impacted by the following: Reduction in circuit loadings. Reduction in MW output of other generators. Bigger generating units displaced off by PV generation (oscillations caused by big generators are reduced). Faster active power recovery (fault ride through capability) of PV inverters. Reduction in net system load (due to embedded PV generation). PV generation not contributing or joining in with power oscillations between the rotating machines on the system. The effect of oscillations between the synchronous machines becoming less significant because of increased PV injection levels. Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 67

77 Section 6 Recommendations 6 RECOMMENDATIONS 6.1 Future work The dynamic studies and simulations carried out in these studies were based on a number of assumptions, as summarised in section 3.1. There are number of possible ways to improve the outcome or the reliability of the outcome from the simulations. The result of the transient stability studies performed for this project are heavily dependent on the dynamic behaviour and control system modelling of PV inverters. If the type of design of inverters being installed changes significantly, the results represented here will need to be reviewed for the new inverter types. The dynamic behaviour of conventional generators, HVDC and STATCOMs or Static Var Compensators (SVCs) is also important but is considered less likely to change. Another important factor is PV inverter distribution across New Zealand. A summary of information used for generator base cases for the studies was mentioned in subsection As explained, derivation of PV generation levels is based on assumptions and a normalisation process. Any deviations or inaccuracies in the predictions used in derivation of solar PV injection levels would have had a noticeable if not significant impact on the dynamic simulations performed for this study. In addition, any differences in the way PV generation has displaced conventional generation in each case (e.g., which generators are dispatched off and how the power output of other generators were increased or decreased) would have also affected important factors like power-flow and the system dynamics. The subsections below describe other areas that may be improved or modified in future PV studies to make accurate predictions regarding the impact of roof-top installed solar inverters on the transient stability of the New Zealand power system Dynamic load modelling For assessing transient stability of this project, default constant load models were used (100% current for active power P and 100% impedance for reactive power Q). Load dynamics have a substantial effect on power system dynamics analysis. Load models have been found to be essential in obtaining adequate accuracy on some power system studies where issues such as motor re-acceleration are important. Power system loads can have dynamic characteristics that may include on-load tap changing transformers, induction motors, thermostats, feeder voltage regulators, voltage controlled feeder capacitors etc. Modelling induction motor loads, especially in the Auckland area and other areas where irrigation loads are prominently present, can affect the power system dynamics to a considerable degree during and after faults. For example, during system faults, (induction) motor type loads such as irrigation tend to draw large amounts of reactive power as they try to re-accelerate. The fault recovery of these units can be slow, thereby Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 68

78 Section 6 Recommendations causing severe voltage dips in the low to medium voltage systems, which would also depress the voltage of the HV transmission system. Consequently, the combination effect of all those motor loads responses could also significantly change generator and PV inverter behaviours. This means the transient stability of the power system is likely to be affected. Conventional generation will generate reactive power automatically if required by the grid (controlled by their voltage regulators and excitation control systems). However, inverters may or may not respond appropriately and may well be limited by current limits if they are at maximum real power outputs. Conversely the need for dynamic modelling of loads is most important where post fault voltage recovery is in doubt, such as at times of high system load and high power transfers towards the loads. These are not the conditions where high PV generation will exist, as the New Zealand peak loads tend to be in the evenings. When high PV generation is running, the transmission lines are expected to be very lightly loaded and will be generating substantial MVARs. It is a recommendation that this area requires some investigation, and that some sensitivity to the load modelling is looked at in the future PV modelling The dynamic behaviour of the three types of inverter technologies (A, B and C) was described in detail in subsection Currently, there is insufficient or robust information available as to the types of inverters that are already in use in New Zealand. For this study, the three types were mainly determined by the inverters known to be widely used in Auckland. This was then assumed for the other regions in North and South Island. Even though there has been a significant increase in uptake of PV generation in New Zealand since 2012, the total installed capacity of roof-installed solar is currently very low (see Figure 68 for the trend in uptake). The assumptions used to model PV are critical to the dynamic behaviour seen in the studies. Therefore, when more robust and accurate data is available in the future, especially in regard to the solar technologies used, PV penetration levels at various regions and grid exit points and inverter connection compliance to connection standards such as AS/NZS :2015, an effort must be made to incorporate the new data into this type of study. Care must be taken to model PV responses as accurately as possible to the responses of actual PV inverters in use. Most importantly, PV responses must be modelled as accurately as possible to the responses of the PV inverters actually in use. For transmission level transient studies, we consider it is acceptable to use aggregated inverter models at the grid exit points. This assessment is based on the nature of their dynamic responses; the non-oscillatory response suggests that an aggregated GXP model will be the worst-case scenario where all PV generation sees the fault the same. Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 69

79 Section 6 Recommendations However, this modelling may need to be investigated further and confirmed as the amount of installed PV increases Critical clearing time Figure 68: Solar PV uptake in New Zealand to May 2017 [11] Critical clearing time is the maximum time during which a disturbance can be applied before one or more generating units lose synchronism with the power system. The critical clearing time for a fault will depend on several factors such as: the fault type, the location of the fault and the capability and characteristics of the in-service generating units. Calculation of critical clearing times was considered outside the scope of this study. However, in later work critical clearing time could be calculated for various PV penetration levels to provide another relative measure of the power system transient performance. This measure may be more robust for the changes in generation patterns that can occur. Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 70

80 Appendix A1: Case details A1 CASE DETAILS A1.1 ON/OFF status of North Island conventional generators Table 10: Statuses of North Island conventional generators in various power-flow cases used (MW max descending order) Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 71

81 Appendix A1: Case details A1.2 ON/OFF Status of South Island conventional generators Table 11: Statuses of South Island conventional generators in various power-flow cases used (MW max descending order) Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 72

82 Appendix A1: Case details A1.3 MW and Mvar output of North Island conventional generators Table 12: MW and MVAR output of North Island generation in the power-flow cases Note: This table should be referred to in conjunction with A1.1 Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 73

83 Appendix A1: Case details A1.4 MW and Mvar output of South Island conventional generators Table 13: MW and MVAR output of South Island generation in the power-flow cases Note: This table should be referred to in conjunction with A1.2 Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 74

84 Appendix A2: The PV models used A2 THE PV MODELS USED Figure 69: Frequency control component of inverter type C dynamic model (simplified) Figure 70: Current limiting logic modelled for inverter type A Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 75

85 Appendix A2: The PV models used Figure 71: Simplified dynamic model control block diagram of Volt-Var component of inverter type A Figure 72: Simplified dynamic model control block diagram of Power Factor component of inverter type A Figure 73: Dynamic model control block diagram of voltage tripping, frequency tripping and reconnect settings component of inverter type C Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 76

86 Appendix A3: Transient responses A3 TRANSIENT RESPONSES A3.1 Rotor angle of North Island generators Figure 74: Rotor angle response of North Island generators. Ref gen. Glenbrook G1 (winter Tuesday, Wairakei fault) Figure 75: Rotor angle response of North Island generators. Ref gen. Glenbrook G1 (winter Saturday, Wairakei fault) Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 77

87 Appendix A3: Transient responses Figure 76: Rotor angle response of North Island generators. Ref gen. Glenbrook G1 (summer Sunday, Wairakei fault) Figure 77: Rotor angle response of North Island generators. Ref gen. Glenbrook G1 (winter Tuesday, Huntly fault) Figure 78: Rotor angle response of North Island generators. Ref gen. Glenbrook G1 (winter Saturday, Huntly fault) Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 78

88 Appendix A3: Transient responses Figure 79: Rotor angle response of North Island generators. Ref gen. Glenbrook G1 (summer Sunday, Huntly fault) A3.2 Rotor angle of South Island generators Figure 80: Rotor angle response of South Island generators. Ref gen. Benmore G1 (winter Tuesday, Clyde fault) Figure 81: Rotor angle response of South Island generators. Ref gen. Benmore G1 (winter Saturday, Clyde fault) Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 79

89 Appendix A3: Transient responses Figure 82: Rotor angle response of South Island generators. Ref gen. Benmore G1 (summer Sunday, Clyde fault) Figure 83: Rotor angle response of South Island generators. Ref gen. Benmore G1 (winter Tuesday, Manapouri fault) Figure 84: Rotor angle response of South Island generators. Ref gen. Benmore G1 (winter Saturday, Manapouri fault) Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 80

90 Appendix A3: Transient responses Figure 85: Rotor angle response of South Island generators. Ref gen. Benmore G1 (summer Sunday, Manapouri fault) Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 81

91 Appendix A4: PV inverter dynamic model response A4 PV INVERTER DYNAMIC MODEL RESPONSE A4.1 Frequency response The three inverter types used for this study provided no additional power as an under-frequency response. This is due to the fact inverters were assumed to be generating at their maximum power in steady-state operation for a given scenario. Figure 86 shows the normalised active power response of all three inverters types which are connected to the New Zealand power system. Inverter type A reduced its active power output to regulate over-frequency with a dead-bank of 0.2 Hz and a relatively low droop; the final frequency rise of around 1 Hz resulted in a reduction to below 50% of output. Inverter type B maintained constant power. Inverter type C showed two modes of over-frequency response: 1. Reduced power output to regulate over-frequency with a dead-band of 1 Hz (modelled). 2. Reduce power output to regulate over-frequency with no dead-band (not included in the models). In addition, inverter type C ramped back to nominal power when system frequency fell below 51 Hz. It was unclear as to whether this was entirely due to the dead-band, or another control loop which resets the power output when system frequency drops below 51 Hz. A conservative approach was taken to model this behaviour which was to include a control loop to ramp power back to maximum output when system frequency drops below 51 Hz. Figure 69 (Appendix A2) shows the control block diagram of inverter C frequency control. Active Power (pu) Time (s) Inverter A Inverter B Inverter C TSAT Bus Frequency (Hz) (normalised) (normalised) (normalised) Frequency (Hz) Figure 86: Inverter over-frequency response Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 82

92 Appendix A4: PV inverter dynamic model response A4.2 Transmission voltage and reactive power response Three tested inverter types A, B and C had three types of selected mode of reactive power response. 1. Volt-Var Mode 2. Constant Power Factor Mode 3. Power-Power Factor Mode In constant Power Factor or Power-Power Factor modes, the inverters maintain their reactive power output in response to transmission voltage variations. Different inverter types displayed different characteristics within these modes of operation. The dynamic models have been set to prioritise active power over reactive power (see Figure 70 Appendix A2). A current limiter was applied to the active and reactive current which is set marginally higher than the initial active current. As such, the actual reactive power response seen from the models are limited. Figure 87 shows the Volt-Var response of the inverter type A. It is important to note that Volt-Var response was generated by setting the inverter active power to zero and by setting the current limit to 1 (e.g., rated MVA current). See Figure 71- Appendix A2, for simplified dynamic control block diagrams of the Volt-Var control mode. Volt-Var response was modelled as four main controls; dead-band (asymmetric up and down), reactive power-voltage droop, reactive power limit at 33% of inverter MVA rating, and PI gains and ramp limiters. The asymmetry of the response (larger response to high transmission voltages) and the variable droop (e.g., the reactive power output per kv voltage change) are of interest for network control. For transmission system studies, and using equivalent grid exit point transformers, these aspects are not critical to the results but for real distribution networks these characteristics may be more critical. Figure 88 shows the constant Power Factor response to a grid frequency change. This shows the interaction of different controls that are active in the inverters, with the subject inverter decreasing power output due to the frequency rise and by modifying reactive power output (if necessary). For these studies, reactive power output of 0 MVAR was assumed, i.e. a power factor of 1.0. In this condition, maintaining the power factor would not change the MVAR output from 0. See Figure 72- Appendix A2 for the simplified control block diagram of power factor control component of inverter type A. Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 83

93 Appendix A4: PV inverter dynamic model response Figure 87: Volt-Var response of inverter type A (voltage measured at the LV side of the inverter transformer winding) Active/Reactiver Power (pu) Time (s) Frequency (Hz) Active Power Reactive Power TSAT Bus Frequency (Hz) Figure 88: Constant Power Factor response of inverter type A A4.3 Reconnecting characteristics All inverter types (A, B and C), displayed different reconnecting characteristics following a frequency or voltage trip. This reconnection is characterised by the reset time and the rate at which the inverter returns to maximum power output. Inverter type A had a reset time of 90 seconds, and returned to its maximum power output with a slow ramp (50% of maximum output over 50 seconds), then a fast ramp (remaining 50% of maximum output over 5 seconds). This characteristic was approximated in the model as a ramp to maximum power output over approximately 40 seconds. Both inverter types B and C had a reset time of 30 seconds and reconnected with a step to maximum power output. Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 84

94 Appendix A4: PV inverter dynamic model response Figure 89 shows the graphical representation of the reconnecting characteristics of inverter types A, B and C for voltage dip in terminal voltage. As noted previously, the ability of the PV inverters to ride through the normal system faults without disconnecting is critical and consequently inverters were modelled as complying with the New Zealand standards in this respect. The connection behaviour was therefore not required to be modelled in the studies presented here. The power recovery characteristics following a short duration block of the inverter (due to low transmission voltages during a fault) was modelled. 1.2 Active Power / Terminal Voltage (pu) Time (s) Inverter A Terminal Voltage Inverter B Terminal Voltage Inverter C Terminal Voltage Inverter A Active Power Inverter B Active Power Inverter C Active Power Figure 89: Reconnecting characteristics of inverter types A, B and C for voltage dip in terminal voltage Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 85

95 Appendix A5: Transient stability theory A5 TRANSIENT STABILITY THEORY With electric power systems, the change in electrical torque of a synchronous machine following a perturbation can be resolved into two components: TT SS δ component of torque change in phase with the rotor angle. TT DD ω component of torque in phase with the speed deviation. If change in electrical torque is TT ee, the relationship is given by Equation 3 [4]. Where: TT SS = synchronising torque coefficient δδ = synchronising torque TT DD = damping torque coefficient ωω = damping torque Equation 3 TT ee = TT SS δδ + TT DD ωω Another important characteristic required to understand transient stability in general is the relationship between interchange power and angular positions of the rotors of synchronous machines. Consider a two-bus system where the generator is connected via a transmission line to a motor at the other end (see Figure 90). The relationship between power transfer and voltages is given by Figure 91 [4]. Power transferred between the two generators depends mainly on the angle difference between them. If Equation 4 is plotted as a function of angle it can be seen how the maximum power transfer between the two generators occurs when the angle difference is 90 degrees, as shown in Equation 5. Figure 90: Two generator network with reactance between them [3] Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 86

96 Appendix A5: Transient stability theory Figure 91: Power-flow between two buses based on the angle difference between them [3] Although the maximum power interchange is at 90 degrees, a power system is rarely operated at that point. The main reason is that the power system in general is dynamic and quantities are constantly varying. Where: Equation 4 PP ee = VV 11VV 22 ssssss δδ XX PP ee =electrical power transferred from the generator to the motor VV 1 = generator voltage VV 2 =motor voltage The power system consists of number of synchronous machines operating synchronously under all operating conditions. During any disturbance, the rotor of a generator decelerates or accelerates with respect to another, creating a relative motion. If the oscillation is stable, the rotors of all machines will achieve the same steady speed after the oscillation, and the generators are transiently stable. This swing of the rotor of synchronous machines is explained by Equation 5. The solution of the swing equation shows how the rotor angle changes in respect to time following a disturbance. The plot of δ against time t is called the Swing Curve. Once the swing curve is known, the stability of the system can be assessed. Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 87

97 Appendix A5: Transient stability theory Where: H = per unit inertia constant ωω 0 = rated angular velocity of the rotor δδ = rotor angle in electrical radians t= time in seconds PP mm = per unit mechanical power input Equation 5 22HH. dd22 δδ = PP ωω 00 ddtt 22 mm PP mmmmmm ssssss δδ KK DD ( 11. dddd ) ωω 00 dddd PP mmmmmm = per unit maximum electrical power output KK DD = damping coefficient Physically, the generator accelerates from its initial steady angle to the new angle required for the machine's electrical power conversion (Pe) to match the new mechanical input power, and gains excessive positive kinetic energy in doing so. The rotor angle continues to open out past the required equilibrium value since it arrives at this value with non-zero speed deviation +Δω (excess kinetic energy). This can lead to run-away rotor speed, as the rotor angle reaches a critical value. Beyond this critical angle, the angle continues to increase causing ever greater acceleration in rotor speed until the machine starts to pole slip and lose stability. Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 88

98 Appendix A6: Emerging Energy Programme: plan and outcome strategy A6 EMERGING ENERGY PROGRAMME: PLAN AND OUTCOME STRATEGY Emerging Energy Technologies: Programme Tranche Plan Historic Work 2016/ / /19 Wind Wind Capacity Assessment Solar PV Solar PV Variability Studies Solar PV System Stability Studies Training Battery and Storage Market System, Real Time Operations, Process and People Situational Intelligence Initial Work Battery Storage Trial Invex Stage 1 Situational Intelligence Programme Definition Battery Operations Impact Assessment Situational Intelligence Stage 1 Invex Battery Storage Next Steps Consideration of Economic Options for Investment Situational Intelligence Stage 1 Capex Situational Intelligence Future Phasing Monitoring Progress Against Transmission Tomorrow Future States (ongoing) Work Packages to be Executed with each Emerging Technology Lines Company Data Exchange Review Assessment of Capabilities Load Forecast Review Ancillary Services Review SO Tools Review Policy and Standard Review Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 89

99 Emerging Energy Programme: plan and outcome strategy A6.1 Emerging Energy Technologies Outcome Strategy Map Effect of Solar PV on Transient Stability of the New Zealand Power System Transpower New Zealand Limited. All rights reserved. 90