PES Cook Islands KEMA Grid Study Final Report

Similar documents
Grid Stability Analysis for High Penetration Solar Photovoltaics

Hawai'i Island Planning and Operations MEASURES TO IMPROVE RELIABILITY WITH HIGH DER

CIS-IEEE 2017 Conference Renewable Energy Session Renewable Energy s Impact of Power Systems

THE NECESSITY OF THE 500 KV SYSTEM IN NWE S TRANSMISSION SYSTEM TO MAINTAIN RELIABLE SERVICE TO MONTANA CUSTOMERS

Project #148. Generation Interconnection System Impact Study Report

Interconnection System Impact Study Report Request # GI

The Role of Electricity Storage on the Grid each location requires different requirements

PV Penetration on Island Mini-Grid or Grid!- Niue. King Island Connect 2015

Reactive Power Compensation for Solar Power Plants. Andy Leon IEEE PES Chicago Chapter December 12 th, 2018

TRANSMISSION PLANNING CRITERIA

PLANNING, ELIGIBILITY FOR CONNECTION AND CONNECTION PROCEDURE IN EMBEDDED GENERATION

Supplemental Report on the NCTPC Collaborative Transmission Plan

Gateway South Transmission Project

Distributed Energy Resources

Ancillary Services & Essential Reliability Services

ABB POWER SYSTEMS CONSULTING

PJM Generator Interconnection Request Queue #R60 Robison Park-Convoy 345kV Impact Study September 2008

PID 274 Feasibility Study Report 13.7 MW Distribution Inter-Connection Buras Substation

PV Grid Integration Research in the U.S.

Generator Interconnection Facilities Study For SCE&G Two Combustion Turbine Generators at Hagood

Feasibility Study Report

Electrical grid stability with high wind energy penetration

100 MW Wind Generation Project

Grid Integration Costs: Impact of The IRP Capacity Mix on System Operations

S C ELECTRIC EUROPE LTD. Excellence Through Innovation. Harnessing the Wind. November 2011 Descriptive Bulletin E

Elbert County 500 MW Generation Addition Interconnection Feasibility Study Report OASIS POSTING # GI

CHAPTER 5 FAULT AND HARMONIC ANALYSIS USING PV ARRAY BASED STATCOM

Ukujima Photovoltaic Park 400 MW Stable Integration of a 400MW Photovoltaic Farm into the Japanese Power System Challenges and Chances

Feasibility Study Report

C PER. Center for Advanced Power Engineering Research C PER

Generator Interconnection System Impact Study For

FAULT ANALYSIS OF AN ISLANDED MICRO-GRID WITH DOUBLY FED INDUCTION GENERATOR BASED WIND TURBINE

Presented By: Bob Uluski Electric Power Research Institute. July, 2011

Guideline for Using IEEE 1547 for Solar PV Interconnection Page 1

DFO STATEMENT OF NEED REPORT

The North Carolina solar experience: high penetration of utility-scale DER on the distribution system

Accidental Islanding of Distribution Systems with Multiple Distributed Generation Units of Various Technologies

DEMONSTRATION OF ESSENTIAL RELIABILITY SERVICES BY A 300-MW SOLAR PV POWER PLANT

Feasibility Study Report

Essential Reliability Services From PV Plants

Field Verification and Data Analysis of High PV Penetration Impacts on Distribution Systems

Dynamic Study of Bonaire Island Power System: Model Validation and Project Experience

CHAPTER 3 TRANSIENT STABILITY ENHANCEMENT IN A REAL TIME SYSTEM USING STATCOM

15 Nelson-Marlborough Regional Plan

Service Requested 150 MW, Firm. Table ES.1: Summary Details for TSR #

Feasibility Study for the Q MW Solar Project

Targeted Application of STATCOM Technology in the Distribution Zone

DETOUR GOLD CORPORATION SYSTEM IMPACT ASSESSMENT FOR DETOUR LAKE PROJECT

Workshop on Grid Integration of Variable Renewable Energy: Part 1

Renewable Grid Integration Research in the U.S.

System Impact Study Report

Grid Management Voltage Control Distribution Grid Voltage Regulation with DER. Michael Sheehan, P.E. IREC Pacific Northwest Solar Partnership

EL PASO ELECTRIC COMPANY (EPE) FACILITIES STUDY FOR PROPOSED HVDC TERMINAL INTERCONNECTION AT NEW ARTESIA 345 KV BUS

Small Electrical Systems (Microgrids)

SPS Planning Criteria and Study Methodology

MILESTONE SUMMARY REPORT Project funding provided by customers of Xcel Energy through a grant from the Renewable Development Fund

Interconnection Feasibility Study Report GIP-226-FEAS-R3

MILLIGAN SOLAR PROJECT

The Status of Energy Storage Renewable Energy Depends on It. Pedro C. Elizondo Flex Energy Orlando, FL July 21, 2016

Journal of American Science 2015;11(11) Integration of wind Power Plant on Electrical grid based on PSS/E

GRID CONNECTION ISSUES FOR DISTRIBUTED GENERATION REVIEW & STANDARDS

Falcon-Midway 115 kv Line Uprate Project Report

Georgia Transmission Corporation Georgia Systems Operations Corporation

Y9. GEH2.3: FREEDM Cost Benefit Analysis based on Detailed Utility Circuit Models

Battery Energy Storage System addressing the Power Quality Issue in Grid Connected Wind Energy Conversion System 9/15/2017 1

AMERICAN ELECTRIC POWER 2017 FILING FERC FORM 715 ANNUAL TRANSMISSION PLANNING AND EVALUATION REPORT PART 4 TRANSMISSION PLANNING RELIABILITY CRITERIA

Project #94. Generation Interconnection System Impact Study Report Revision

A Case Study on Aggregate Load Modeling in Transient Stability Studies

Evaluation of the Performance of Back-to-Back HVDC Converter and Variable Frequency Transformer for Power Flow Control in a Weak Interconnection

Relative Power Factor Correction costs

Advanced controls of PV for microgrid applications

Energy Security Electrical Islanding Approach and Assessment Tools. Dr. Bill Kramer Senior Research Engineer Distributed Energy Systems Integration

Electric Power Research Institute, USA 2 ABB, USA

Stability Study for the Mt. Olive Hartburg 500 kv Line

Regenerative Utility Simulator for Grid-Tied Inverters

15 Nelson-Marlborough Regional Plan

Session 10 NERC Interconnection Requirements

Essential Reliability Services Engineering the Changing Grid

CUSTOMER/ TWIN ARROWS PROJECT

ABB Microgrids and Energy Storage. Nathan Adams, Director, Technology and Business Development

Operational Planning Study Report. RTA to BCH transfer limit updates For Kitimat 4 Capacitor Banks

RULE 21 GENERATING FACILITY INTERCONNECTION APPLICATION SMUD s Distribution System - (SMUD FORM 2655)

Final Draft Report. Assessment Summary. Hydro One Networks Inc. Longlac TS: Refurbish 115/44 kv, 25/33/ General Description

The Power Potential Project A guide to participating

Q217 Generator Interconnection Project

Impact of Distributed Energy Resources on Transmission System Reliability

ANCILLARY SERVICES WITH VRE (VARIABLE RENEWABLE ENERGY): FOCUS PV

A Cost Benefit Analysis of Faster Transmission System Protection Schemes and Ground Grid Design

An Integrated Grid Path for Solar. Thomas Key, EPRI Senior Technical Executive. ISES Webinar. April 22, 2016

Presentation to the PJM NEMSTF Interconnection Issues. Interconnecting Solar Projects on the PSE&G Distribution System

Interconnection Feasibility Study Report GIP-023-FEAS-R1. Generator Interconnection Request # MW Wind Generating Facility Inverness (L6549), NS

Draft Guideline for the connection of smallscale inverter based distributed generation. EEA Asset Management Forum 22 June 2016 Dr Richard Strahan

Southern Company Interconnection Process. Dexter Lewis Research Engineer Research and Technology Management

To Shift or not to Shift?

Analysis of Grid Connected Solar Farm in ETAP Software

DUKE ENERGY PROGRESS TRANSMISSION SYSTEM PLANNING SUMMARY

DIgSILENT Pacific PowerFactory Technical Seminar

Galapagos San Cristobal Wind Project. VOLT/VAR Optimization Report. Prepared by the General Secretariat

Distributed Energy Resources

Experience on Realizing Smart Grids. IEEE PES conference, Gothenburg

Transcription:

Integrating PV Solar and Wind generation with the TAU electric system Te Aponga Uira O Tumu-Te_Varovaro

Contents 1. Executive summary... 2 2. Introduction... 5 2.1 Proposed wind turbine sites... 6 2.2 Proposed PV sites... 7 2.3 Steady state (spreadsheet analysis)... 7 2.4 Simulation assumptions... 8 2.5 Project scenarios... 9 2.5.1 Night time off-peak load... 9 2.5.2 Day time peak load... 15 2.6 Steady-state conclusions... 21 3. Dynamic simulation results... 22 3.1 Dynamic issues with island power systems... 22 3.1.1 Solar and wind generation inverter voltage and frequency ride-through... 22 3.1.2 Island small system sizes... 24 3.1.3 Technical solutions... 25 3.2 Base case daytime peak load (baseline without RE)... 25 3.3 Base case off-peak load (baseline without RE)... 29 3.4 Night case with wind generation... 33 3.5 Day case with wind and PV generation... 37 3.6 Equipment controls... 41 4. Revised Modeling and Scenarios... 42 4.1 Revised Day case with wind and PV generation... 42 4.1.1 Day case with revised RE Dynamic Stability Analysis... 51 4.2 Revised Night case with wind generation... 56 4.2.1 Night case with revised RE Power Quality Discussion... 63 4.2.2 Night case with revised RE Dynamic Stability Analysis... 63 4.3 Revised Day case with maximum PV generation... 70 4.3.1 Day case with maximum PV Power Quality Discussion... 79 4.3.2 Day case with maximum PV Dynamic Stability Analysis... 79 5. Conclusion... 86 Figures Figure 1: Rarotonga load curve... 5 Figure 2: Potential Wind Farm Sites on Rarotonga... 6 Integrating PV Solar and Wind generation with the TEC electric system i

Contents Figure 3: Standard voltage ride-through limits... 23 Figure 4: Standard frequency ride-through limits... 24 Figure 5: Trip of GS-7 (No fault); Frequency Response... 26 Figure 6: Trip of GS-7 (No fault); Voltage Response... 26 Figure 7: Trip of Cross Line Feeder (No fault); Frequency Response... 27 Figure 8: Trip of Cross Line Feeder (No fault); Voltage Response... 27 Figure 9: Trip Line Feeder with Close-in Fault; Frequency Response... 28 Figure 10: Trip Cross Line Feeder with Close-in Fault; Voltage Response... 28 Figure 11: Trip of GS-7 (No fault); Frequency Response... 29 Figure 12: Trip of GS-7 (No fault); Voltage Response... 30 Figure 13: Trip of Cross Line Feeder (No fault); Frequency Response... 30 Figure 14: Trip of Cross Line Feeder (No fault); Voltage Response... 31 Figure 15: Trip Line Feeder with Close-in Fault; Frequency Response... 32 Figure 16: Trip Cross Line Feeder with Close-in Fault; Voltage Response... 32 Figure 17: Off-Peak with Wind; Loss of GS-7 (No fault); Frequency Response... 33 Figure 18: Off-Peak with Wind; Loss of GS-7 (No fault); Voltage Response... 34 Figure 19: East Coast Feeder fault; Frequency Response... 35 Figure 20: East Coast Feeder fault; Voltage Response... 35 Figure 21: Loss of Largest Wind Resource (No fault); Frequency Response... 36 Figure 22: Loss of Largest Wind Resource (No fault); Voltage Response... 37 Figure 23: Peak with Wind & PV; Loss of GS-7 (No fault); Frequency Response... 38 Figure 24: Peak with Wind & PV; Loss of GS-7; Voltage Response... 38 Figure 25: Seaport Feeder fault; Frequency Response... 39 Figure 26: Seaport Feeder fault; Voltage Response... 39 Figure 27: Loss of Sun City 2000 kw PV (No fault); Frequency Response... 40 Figure 28: Loss of Sun City 2000 kw PV (No fault); Voltage Response... 40 Figure 29: Peak with revised Wind & PV; Loss of GS-7 (No fault); Frequency Response... 52 Figure 30: Peak with revised Wind & PV; Loss of GS-7 (No fault); Voltage Response... 52 Figure 31: Seaport Feeder fault; Frequency Response... 53 Figure 32: Seaport Feeder fault; Voltage Response... 54 Figure 33: Loss of Sun City 2000 kw PV (No fault); Frequency Response... 55 Figure 34: Loss of Sun City 2000 kw PV (No fault); Voltage Response... 55 Figure 35: Off-Peak with revised RE; Loss of GS-7 (No fault); Frequency Response... 63 Figure 36: Off-Peak with revised RE; Loss of GS-7 (No fault); Voltage Response... 64 Integrating PV Solar and Wind generation with the TEC electric system ii

Contents Figure 37: West Coast Feeder fault; Frequency Response... 65 Figure 38: West Coast Feeder fault; Voltage Response... 65 Figure 39: Loss of Raemaru 2000 kw Wind (No fault); Frequency Response... 67 Figure 40: Loss of Raemaru 2000 kw Wind (No fault); Voltage Response... 67 Figure 41: Loss of Raemaru 2000 kw Wind (No fault); Diesel Generator Real Power Output... 68 Figure 42: Loss of Raemaru 2000 kw Wind (No fault); Diesel Generator Reactive Power Output... 69 Figure 43: Loss of Raemaru 2000 kw Wind (No fault); Diesel Generator Rotations Per Minute... 69 Figure 44: Voltage Flicker curve from IEEE 141 & 519... 79 Figure 45: Maximum PV; Loss of GS-7 (No fault); Frequency Response... 80 Figure 46: Maximum PV; Loss of GS-7 (No fault); Voltage Response... 80 Figure 47: Maximum PV; Airport Feeder fault; Frequency Response... 81 Figure 48: Maximum PV; Airport Feeder fault; Voltage Response... 81 Figure 49: Maximum PV; Loss of 75% PV Generation (3358 kw) (No fault); Frequency Response... Error! Bookmark not defined. Figure 50: Maximum PV; Loss of 75% PV Generation (3358 kw) (No fault); Voltage Response... Error! Bookmark not defined. Figure 51: Maximum PV; Loss of 75% PV Generation (3358 kw) (No fault); Diesel Generator Real Power Output... 84 Figure 52: Maximum PV; Loss of 75% PV Generation (3358 kw) (No fault); Diesel Generator Reactive Power Output... 85 Figure 53: Maximum PV; Loss of 75% PV Generation (3358 kw) (No fault); Diesel Generator Rotations Per Minute... 85 Tables Table 1: Abbreviations... 1 Table 2: Proven feeder and total island renewable generation limits... Error! Bookmark not defined. Table 3: Potential wind turbine capacity per site 3... 7 Table 4: Theoretical feeder and total island renewable generation limits (absent energy storage)... 7 Table 5: Renewable Resources Summary... 9 Table 6: Night time base case system summary (pre RE)... 9 Table 7: Night time case with wind turbines system summary... 10 Table 8: Night time case with wind turbines feeder and transformer loading summary... 10 Table 9: Day time base case system summary (pre RE)... 15 Table 10: Day time case with wind turbines and PV inverters system summary... 16 Integrating PV Solar and Wind generation with the TEC electric system iii

Contents Table 11: Day time case w/wind turbines & PV inverters feeder and transformer loading summary... 16 Table 12: Standard voltage ride-through settings... 24 Table 13: Potential Large-scale PV Installations... 42 Table 14: Renewable Resources Summary... 44 Table 15: Day time peak case with revised RE system summary... 44 Table 16: Day time peak case with revised RE generation summary... 45 Table 17: Day time peak case with revised RE feeder and transformer loading summary... 45 Table 18: Night time off-peak case with revised RE system summary... 56 Table 19: Night time off-peak case with revised RE generation summary... 56 Table 20: Night time off-peak case with revised RE feeder and transformer loading summary... 57 Table 21: Potential Large-scale PV Installations... 70 Table 22: Maximum PV Renewable Resources Summary... 71 Table 23: Day time peak case with maximum PV system summary... 72 Table 24: Day time peak case with maximum PV generation summary... 72 Table 25: Day time peak case with maximum PV feeder and transformer loading summary... 73 Table 26: Renewable Energy Capacity Factors... 86 Table 27: Annual Energy and Load Factor... 87 Integrating PV Solar and Wind generation with the TEC electric system iv

Table 1: Abbreviations Deg Degree DRG distributed renewable generation kw kilo Watt a thousand Watts Pf Power factor PQG Generator supplying real (P) and reactive (Q) power PV Photo-voltaic (solar) plant RMU Ring main unit SW Generator swing bus in power flow TAU Te Aponga Uira O Tumu-Te Varovaro Var A unit (Volt-Ampere) of reactive power Vpu Voltage per-unit (relative to nominal voltage) W Watt, a unit (Volt-Ampere) of real power Integrating PV Solar and Wind generation with the TEC electric system 1

1. Executive Summary The government of the Cook Islands has announced goals to achieve 50% of electric energy requirements on each island from renewable energy (RE) by 2015 and 100% by 2020. In a previous study KEMA Australia analyzed the economic viability of potential scenarios to achieve these ambitious RE goals. 1 This will entail a significant change in system operation and technical study is needed to identify the practical limit on wind and PV generation that can be installed on the island while maintaining system reliability and power quality. The economic viability study did not include assessment of the electrical performance of the Rarotonga system with such RE scenarios. In the current study DNV KEMA was asked to evaluate the maximum practical amount of photovoltaic (PV) and wind generation that could reliably be added to the Te Aponga Uira O Tumu-Te Varovaro (TAU) electrical network located on Rarotonga, in the Cook Islands. The analysis required power system simulation. For these simulations DNV KEMA used the same network model that was used for a recent efficiency (loss reduction) study it performed for TAU and the Pacific Power Authority. 2 The new study examines system performance with RE deployment levels slightly below the 2015 target of 50%. The maximum practical level of RE deployment was analyzed at both the island level and the feeder level. Steady-state and dynamic simulations of the island grid were performed to analyze power flows, voltages and stability performance. A number of on-peak and off-peak scenarios were simulated based on the potential RE project sites and sizes that have been identified. Storage options were not modeled in the current study. However, it was observed that due to power quality concerns, the maximum level of PV deployment may depend upon installation of local storage capability at larger PV sites currently being considered. Table 2 summarizes the levels of PV and wind output that can be accommodated on the current system. 1. TAU Final Renewable Economic Viability Study, prepared by KEMA Australia, September 2012. 2. KEMA, Quantification of the Power System Energy Losses in South Pacific Utilities, Te Aponga Uira O Tumu -Te- Varovaro, Cook Islands, May 2012. Integrating PV Solar and Wind generation with the TEC electric system 2

Table 2: Proven feeder level and Total Island RE generation limits Rarotonga, Cook Islands Daytime (PV), no storage Daytime (PV), local storage Nighttime (Wind), no storage Feeder level (non-coincident) 3 Airport Feeder 1500-2000 kw 2000 kw 500 kw Avarua City Feeder n/a n/a n/a East Coast Feeder 500 kw 500 kw 100 kw Seaport Feeder 2000 kw 3000 kw n/a Cross Line Feeder 300 kw 300 kw 310 kw 4 West Coast Feeder 500 kw 500 kw 2000 kw 5 Total island (coincident) 6 RE generation limit 7 1500-2000 kw 4480 kw 2000 kw Note - the limits shown in Table 2 exclude small behind the meter PV installations. The analysis was divided into two parts - steady state analysis and dynamic analysis: 1. The steady state analysis limited the amount of additional PV and wind generation on a given feeder to the output that would load feeder conductors up to their continuous thermal loading capability, and assumes that feeder protective relaying can handle reverse power flow back to the generator station. A preliminary estimate of the maximum RE generation limit for the island was initially developed by subtracting Avatiu Valley Power Station diesel generation output from the island peak and off peak load, assuming sufficient diesel generation is dispatched to cover for the loss of the largest diesel unit and base case dispatch set in the range of 25-50% of machine rating in order to allow the diesels to regulate frequency for RE output swings in either direction (up or down). Based on these steady state criteria, it was 3 Not all of the individual feeder RE limits are coincident with each other and/or with the island RE limit. 4 Proposed wind projects on the feeder were reduced to 100 kw at West Wigmores and 50 kw at Muri. 5 Proposed wind project sizes on the feeder are either 1,100 kw or 3,000 kw. Grid analysis was done at a mid-range value of 2,000 kw and had acceptable performance. No analysis was done for a 3,000 kw project in the absence of storage since there is insufficient night-time demand to accept this level of wind generation. 6 This represents the maximum coincident RE output on the island as demonstrated by the grid analysis. 7 The maximum daytime combination of PV/wind generation modeled was 3678 kw of PV plus 700 kw of wind. However, this combination might require installation of local storage capability at the larger PV sites. Integrating PV Solar and Wind generation with the TEC electric system 3

initially estimated that a maximum of 3,300 kw of RE could be added to the existing island system. The detailed grid analysis was then performed to refine these estimates. 2. The dynamic analysis started with the steady state parameters and modeled several onpeak and off-peak scenarios. Each case examined grid voltage, frequency and other performance variables for critical disturbances such as diesel generator tripping, feeder faults and loss of RE generation. The study identified a potential limitation on PV deployment due to flicker (power quality) concerns. It was concluded that this concern could be mitigated in a number of ways, including: 1. Limiting the total installed PV capacity in the northwest portion of the island to 1,500-2,000 kw, or 2. Installing PV up to levels shown in the middle column of Table 2 together with a local battery storage system at each of the larger PV projects (e.g., projects larger than 500 kw) with fastacting control logic that can provide smoothing of rapid PV output swings and keep grid voltage fluctuations at 2% or less during passing cloud fronts, or 3. Conduct more detailed study of the proposed PV project locations vs. cloud patterns over the island to determine if there is enough non-coincidence in cloud cover at these locations to smooth out the resulting grid voltage swings and meet acceptable flicker levels. Integrating PV Solar and Wind generation with the TEC electric system 4

0:30 1:30 2:30 3:30 4:30 5:30 6:30 7:30 8:30 9:30 10:30 11:30 12:30 13:30 14:30 15:30 16:30 17:30 18:30 19:30 20:30 21:30 22:30 23:30 Total Load (kw) DNV KEMA Energy & Sustainability 2. Introduction Rarotonga, Cook Islands has an 11 kv electric system serving a peak load of 4,830 kw in 2011. The electric distribution system is entirely underground with 11 kv primary and 415 V secondary systems. Te Aponga Uira O Tumu-Te_Varovaro (TAU) is considering installing as much Solar and Wind generation as the island system can accommodate. To that extent, this study will determine the amount of distributed renewable generation (DRG) that the Rarotonga 11 kv system can allow before implementation of other measures (such as energy storage) becomes necessary. Two simulations will be completed during the first phase of the project. Figure 1 is a 24-hour load curve from historical loading data collected from TAU for the peak loading day in 2011. This load curve will be the basis of these simulations. Figure 1: Rarotonga load curve 8 5000 4500 4000 3500 3000 2500 2000 Rarotonga Daily Load Curve HOUR At present there are only small PV installations on the island and they are connected behind the meter. There are presently no wind generator turbines on Rarotonga. TAU wishes to achieve a level of 50% of the island load served by renewable resources by 2015 9. This will entail a significant change in system operation and technical study is needed to identify the maximum potential wind and PV generation that can be installed on the island while maintaining system reliability and power quality. 8 Source: TAU Daily Operator Log sheets, June 22 nd, 2011 9 Source: TAU Final Renewable Economic Viability Study, prepared by KEMA Australia, September 2012. Integrating PV Solar and Wind generation with the TEC electric system 5

2.1 Proposed wind turbine sites Figure 2 below shows the 5 potential sites for wind generation and Table 2 summarizes the turbine types and sizes proposed for each location. Figure 2: Potential Wind Farm Sites on Rarotonga 10 10. Source: Jessica McMahon e-mail; November 22, 2012 Integrating PV Solar and Wind generation with the TEC electric system 6

Table 2: Potential wind turbine capacity per site 3 Site Scenario Turbine Model No. Turbines Total Capacity (kw) Matevera 1 Windflow 500 2 1000 Muri 1 Generic 1 50 West Wigmores 1 Windflow 500 1 500 Raemaru 1 Vergnet GEV HP 3 3000 2 Vergnet GEV MP 4 1100 Hospital 1 Vergnet GEV HP 1 1000 2 Vergnet GEV MP 2 550 2.2 Proposed PV sites There are plans to install a Sun City 2000 kw PV project ( Sun City ) near the Rarotonga Airport. This PV site is close to both the feeder north of the airport along the coast and the feeder to the south of the airport. For modeling purposes, KEMA interconnected the Sun City project to the Seaport Feeder. 2.3 Steady state (spreadsheet analysis) DNV KEMA first performed a tabletop (spreadsheet) analysis in order to estimate the maximum potential RE amounts that it might be possible to install on the overall Rarotonga system as well as on individual feeders. The feeder RE limits shown in assume that existing protective relaying on the feeders is currently capable of accepting back feed (power flowing backward from the feeder into the substation bus), or can be modified to accommodate back feed. This spreadsheet analysis resulted in the initial estimate of installed RE capacity limits as shown in Table 2: Table 3: Theoretical feeder and total island renewable generation limits (absent energy storage) Rarotonga, Cook Islands Day Night Feeder 1 Airport (A) 4383 kw 4049 kw Feeder 2 Avarua City (AC) 6384 kw 6002 kw Feeder 3 Cross Line (CL) 6813 kw 6217 kw Feeder 4 East Coast (EC) 6393 kw 6007 kw Integrating PV Solar and Wind generation with the TEC electric system 7

Feeder 5 Seaport (S) 6287 kw 5954 kw Feeder 6 West Coast (WC) 6459 kw 6040 kw Island max 3330 kw 1400 kw Both the feeder limits and the island maximum numbers are subject to the following simulation results which DNV KEMA performed using the EasyPower software package. The simulation assumptions, methodology and results are described in the following sections. 2.4 Simulation assumptions Following are the assumptions used in the power flow model for the initial daytime and nighttime RE scenarios as described in more detail the following section: Model from previous DNV KEMA Cook Islands loss study used as the base case. Load scaled down to 2,400 kw off-peak (NIGHT) and 4,830 kw on-peak (DAY) load. Two diesel generators were assumed to be running in the both scenarios; 2 units online in the DAY base case dispatched at 1,000 kw and 609 kw, respectively; and 2 units online in the NIGHT base case dispatched at 500 kw and 614 kw. At least one diesel unit is needed to provide a regulated voltage and frequency source for the island. Running two diesel units allows either diesel unit to continue regulating voltage and frequency if the other unit trips off line. This is typically referred to as an n-1 resource criterion. One PV site was assumed near Rarotonga Airport (Seaport Feeder) rated 2000 kw as shown below in Table 4. Three wind generation sites selected from the five proposed sites as shown in Table 4, based on the sites with closest proximity to existing feeders. Wind Turbine output is assumed to run fairly close to rated project capacity over most of the daily 24 hour period, but de-rated by about 40-50% in late morning through early afternoon. Integrating PV Solar and Wind generation with the TEC electric system 8

Table 4: Renewable Resources Summary Site DRG Type Capacity (kw) Assumed Output (kw) Matevera Wind 1,000 500 Hospital Wind 1,000 500 West Wigmores Wind 500 300 Airport PV (daytime only) 2,000 2,000 2.5 Project scenarios During these simulations, two scenarios were modeled (off-peak night & on-peak day). These cases were used to test the electric system s stability for the maximum on-peak and off-peak RE deployment scenario. Three wind generation locations were chosen based on their proposed sizes and accessibility to the existing power system and the 2000 kw Sun City solar installation was assumed near the Rarotonga Airport. In order to establish a baseline for the island s dynamic performance, the following pre-re dispatch scenarios were modeled. 2.5.1 Night time off-peak load The following table shows the night time base case load and generation: Table 5: Night time base case system summary (pre RE) Total MW Mvar MVA pf Generation in System 2.421 0.304 2.440 0.992 GS-1 0.421 0.104 0.434 0.971 GS-2 1.000 0.100 1.005 0.995 GS-7 1.000 0.100 1.005 0.995 Load in System 2.400 1.163 2.667 0.900 Shunt Load in System 0 0 Losses in System 0.021-0.858 Check of Balance 0 0 Integrating PV Solar and Wind generation with the TEC electric system 9

For the first scenario, wind generators were added at the Hospital, Metevera and West Wigmores sites as described above. The wind turbines were connected to the Airport, East Coast and Cross Line Feeders, respectively. Table 6: Night time case with wind turbines system summary Name Generator Type Solution Rated MVA MW Mvar MVA Pf Vpu Deg GS-1 Sw 2.000 0.614-0.016 0.614-1.000 1.020 0 GS-7 PQG 2.700 0.500 0.100 0.510 0.707 1.018-1.08 INV_WND_HOS PQG 1.500 0.500 0.070 0.505 0.990 1.025-0.81 INV_WND_MET PQG 1.500 0.500 0.075 0.506 0.989 1.025-0.89 INV_WND_WIG PQG 0.750 0.300 0.042 0.303 0.990 1.015-0.90 In the night time case, a total of 1,300 kw of wind generation was modeled taking the conventional generation (diesel) output down to 1,114 kw. With this maximum penetration of wind generation, the system experienced no feeder or transformer overloads as shown in Table 7: Table 7: Night time case with wind turbines feeder and transformer loading summary From Bus Feeder To Bus Branch/Xfmr Name Rated Amps Load Amps % A-107 A-108 A-2 295 17 5.80% A-107 A-PC1 A-3 195 7.8 4.00% A-108 A-108_A A-4 195 1.8 0.90% A-108 A-108_B A-7 295 21.1 7.20% A-108 A-129_A A-5 195 0.8 0.40% A-108_B A-109 A-8 295 23.6 8.00% A-109 A-303 A-9 295 24.4 8.30% A-129 A-129_A A-6 295 0.9 0.30% A-PC1 A-PC3 A-10 195 4.4 2.30% A-PC1 A-PC2 A-13 145 0.8 0.60% A-PC3 A-PC3_B A-11 145 1.8 1.20% Integrating PV Solar and Wind generation with the TEC electric system 10

From Bus To Bus Branch/Xfmr Name Rated Amps Load Amps % A-PC3_B A-PC3_C A-12 195 0.9 0.40% AC-101 AC-103 AC-4 295 18.8 6.40% AC-102 AC-102_A AC-3 295 0 0.00% AC-103 AC-126 AC-6 295 11.7 4.00% AC-103 AC-102 AC-5 230 2.6 1.10% AC-112 AC-112_B AC-13 295 5.7 1.90% AC-112_A AC-112 AC-12 295 6.6 2.20% AC-112_B AC-130_A AC-14 230 5.7 2.50% AC-118 AC-132_A AC-18 230 1.7 0.70% AC-126 AC-112_A AC-11 230 6.5 2.80% AC-126 AC-135_A AC-7 230 0.8 0.40% AC-130 AC-130_B AC-16 295 4.3 1.50% AC-130_A AC-130 AC-15 295 5.7 1.90% AC-130_B AC-118 AC-17 230 4.3 1.90% AC-132 AC-132_B AC-16B 230 0.1 0.00% AC-132_A AC-132 AC-19 230 1.7 0.80% AC-135 AC-135_B AC-9 295 0 0.00% AC-135_A AC-135 AC-8 295 0.9 0.30% BUS-1 EC-119 EC-1 295 7.8 2.60% BUS-1 AC-101 AC-1 295 20.5 6.90% BUS-2 BUS-1 PS BUS TIE 541 18.4 3.40% BUS-2 WC-110 WC-1 295 21.3 7.20% BUS-2 CL-128 CL-1 295 14.4 4.90% BUS-2 A-107 A-1 195 8.8 4.50% BUS-2 S-105 S-1 295 17.5 5.90% CL-128 CL-400_A CL-3 295 13.6 4.60% CL-128 CL-128_A CL-2 195 0.9 0.50% CL-205 CL-213 CL-40 295 3.4 1.20% CL-205 CL-215 CL-39 189 2.3 1.20% CL-206 CL-220_A CL-32 195 11.6 5.90% CL-207 CL-208_A CL-22 195 15.3 7.90% Integrating PV Solar and Wind generation with the TEC electric system 11

From Bus To Bus Branch/Xfmr Name Rated Amps Load Amps % CL-208 CL-208_B CL-24 295 14.5 4.90% CL-208_A CL-208 CL-23 295 15.4 5.20% CL-208_B CL-219_A CL-25 195 14.5 7.50% CL-209 CL-214_A CL-18 195 17.6 9.00% CL-210 CL-221_A CL-14 195 5.5 2.80% CL-210 CL-313 CL-7 195 6.3 3.20% CL-211 CL-211_B CL-30 295 12.9 4.40% CL-211_A CL-211 CL-29 295 13.8 4.70% CL-211_B CL-206 CL-31 195 12.9 6.60% CL-213 CL-212 CL-41 295 1.7 0.60% CL-214 CL-214_B CL-20 295 16.9 5.70% CL-214_A CL-214 CL-19 295 17.7 6.00% CL-214_B CL-207 CL-21 295 16.9 5.70% CL-218 CL-218_B CL-37 230 8.2 3.60% CL-218_A CL-218 CL-36 230 9.1 4.00% CL-218_B CL-205 CL-38 195 8.3 4.20% CL-219 CL-219_B CL-27 295 13.7 4.70% CL-219_A CL-219 CL-26 295 14.6 5.00% CL-219_B CL-211_A CL-28 195 13.7 7.00% CL-220 CL-220_B CL-34 295 9 3.10% CL-220_A CL-220 CL-33 295 11.7 4.00% CL-220_B CL-218_A CL-35 195 9 4.60% CL-221_A CL-221 CL-15 295 5.6 1.90% CL-221_B CL-209 CL-17 195 19.3 9.90% CL-221_B CL-221 CL-16 295 19.3 6.50% CL-312 CL-312_C CL-12 295 0 0.00% CL-312 CL-312_A CL-9 295 5.8 1.90% CL-312_B CL-312 CL-10 295 1.3 0.40% CL-312_B CL-311 CL-11 195 1.3 0.70% CL-313 CL-312_A CL-8 195 5.6 2.80% CL-400_A CL-400 CL-4 195 14.2 7.30% Integrating PV Solar and Wind generation with the TEC electric system 12

From Bus To Bus Branch/Xfmr Name Rated Amps Load Amps % CL-400_B CL-210 CL-6 295 11.6 3.90% CL-400_B CL-400 CL-5 195 11.6 6.00% EC-119 EC-120 EC-2 295 7.3 2.50% EC-120 EC-121 EC-4 295 11.6 3.90% EC-120 EC-131 EC-3 195 2.7 1.40% EC-121 EC-133_A EC-5 195 14 7.20% EC-122 EC-123_A EC-9 195 16.4 8.40% EC-123 EC-123_B EC-114 295 18.9 6.40% EC-123_A EC-123 EC-10 295 16.3 5.50% EC-123_B EC-124 EC-12 195 18.9 9.70% EC-124 EC-134_A EC-13 195 19.7 10.10% EC-133 EC-133_B EC-7 295 14.7 5.00% EC-133_A EC-133 EC-6 295 13.9 4.70% EC-133_B EC-122 EC-8 195 14.7 7.50% EC-134 EC-134_B EC-15 295 5.9 2.00% EC-134_A EC-134 EC-14 295 6.7 2.30% EC-134_B EC-202 EC-16 195 5.8 3.00% EC-202 EC-216_A EC-17 195 4.2 2.20% EC-203 EC-204 EC-24 295 1.3 0.40% EC-216 EC-216_B EC-19 295 3.5 1.20% EC-216_A EC-216 EC-18 295 4.2 1.40% EC-216_B EC-217_A EC-20 195 3.4 1.80% EC-217 EC-217_B EC-22 295 2.6 0.90% EC-217_A EC-217 EC-21 295 3.4 1.20% EC-217_B EC-203 EC-23 195 2.6 1.30% S-105 S-125 S-2 195 15.6 8.00% S-111 S-111_B S-17 295 0 0.00% S-111_A S-111 S-16 295 1.8 0.60% S-113 S-116_A S-12 230 4.3 1.90% S-114 S-117_A S-8 230 6.7 2.90% S-115 S-115_B S-6 230 8.4 3.60% Integrating PV Solar and Wind generation with the TEC electric system 13

From Bus To Bus Branch/Xfmr Name Rated Amps Load Amps % S-115_A S-115 S-5 230 11 4.80% S-115_B S-114 S-7 230 8.4 3.60% S-116 S-116_B S-14 295 1.7 0.60% S-116_A S-116 S-13 295 4.3 1.50% S-116_B S-111_A S-15 230 1.7 0.70% S-117 S-117_B S-10 295 5.9 2.00% S-117_A S-117 S-9 295 6.8 2.30% S-117_B S-113 S-11 230 6 2.60% S-125 S-127 S-3 195 2.6 1.40% S-125 S-115_A S-4 230 10.9 4.70% WC-110 WC-302_A WC-3 230 20.2 8.80% WC-110 WC-301 WC-2 195 0.3 0.10% WC-302 WC-302_A WC-4 295 20.2 6.90% WC-302 WC-315 WC-5 295 18.5 6.30% WC-304 WC-305_A WC-9 195 14.5 7.40% WC-304 WC-318 WC-7 295 3.3 1.10% WC-305 WC-305_B WC-11 295 12 4.10% WC-305_A WC-305 WC-10 295 14.5 4.90% WC-305_B WC-306 WC-12 195 12 6.10% WC-306 WC-307 WC-13 195 9.9 5.10% WC-307 WC-308 WC-15 195 7.4 3.80% WC-307 WC-316 WC-14 295 0.8 0.30% WC-308 WC-309 WC-16 195 5.4 2.80% WC-309 WC-321_A WC-17 295 1.8 0.60% WC-309 WC-314_A WC-21 195 2.1 1.10% WC-314 WC-314_B WC-23 230 1.3 0.60% WC-314_A WC-314 WC-22 230 2.2 1.00% WC-314_B WC-310 WC-24 195 1.3 0.70% WC-315 WC-304 WC-6 295 17.7 6.00% WC-318 WC-319 WC-8 295 2.6 0.90% WC-321 WC-321_B WC-19 295 1 0.30% Integrating PV Solar and Wind generation with the TEC electric system 14

From Bus To Bus Branch/Xfmr Name Rated Amps Load Amps % WC-321_A WC-321 WC-18 295 1.7 0.60% WC-321_B WC-317 WC-20 295 0.9 0.30% Transformers BUS-1 GS-1 TR-1 131 32 24.50% GS-2 BUS-1 TR-2 131 0 0.00% GS-4 BUS-1 TR-3 39 0 0.00% GS-5 BUS-1 TR-4 39 0 0.00% NZ-GS BUS-1 TR-7 131 0 0.00% 2.5.2 Day time peak load For the peak load scenario load was scaled to 4,830 kw to simulate peak day time conditions. The following table shows a summary of the generation and load in the peak load base case: Table 8: Day time base case system summary (pre RE) Total MW Mvar MVA pf Generation in System 4.906 1.524 5.138 0.955 GS-1 0.406 0.074 0.413 0.984 GS-2 1.000 0.500 1.118 0.894 GS-3 1.000 0.400 1.077 0.928 GS-7 2.000 0.500 2.062 0.970 NZ-GS-1 0.500 0.050 0.502 0.995 Load in System 4.830 2.339 5.367 0.900 Shunt Load in System 0 0 Losses in System 0.077-0.815 Check of Balance 0 0 For the day time case with RE added, a 2000 kw PV site was modeled near the Airport, interconnected with the Seaport Feeder. The same wind generation projects were modeled as described above in the offpeak scenario. Table 9 shows the base case dispatch conditions modeled for the PV project, wind turbines and diesel generators. Integrating PV Solar and Wind generation with the TEC electric system 15

Table 9: Day time case with wind turbines and PV inverters system summary Name Generator Type Solution Rated MVA MW Mvar MVA PF Vpu Deg GS-1 Sw 2.000 0.609 0.565 0.831 0.733 1.020 0 GS-7 PQG 2.700 1.000 0.500 1.118 0.894 1.000-0.94 INV_PVC_1 PQG 2.500 2.000 0.285 2.020 0.990 1.022-0.18 INV_WND_HOS PQG 1.500 0.500.0700 0.505 0.990 1.002-0.59 INV_WND_MET PQG 1.500 0.500 0.075 0.506 0.989 0.999-0.69 INV_WND_WIG PQG 0.750 0.300 0.042 0.303 0.990 0.985-0.68 In the day time case, a total of 1,300 kw of wind generation and 2,000 kw of PV generation were modeled taking the conventional generation output down to 1609 kw. With this level of penetration of wind and PV generation the system experienced no feeder or transformer overloads as shown in Table 10. Table 10: Day time case w/wind turbines & PV inverters feeder and transformer loading summary From Bus Feeder To Bus Branch/Xfmr Name Rated Amps Load Amperes % A-107 A-108 A-2 295 8.2 2.80% A-107 A-PC1 A-3 195 16.3 8.30% A-108 A-108_A A-4 195 3.7 1.90% A-108 A-129_A A-5 195 1.7 0.90% A-108 A-108_B A-7 295 16.3 5.50% A-108_B A-109 A-8 295 21.5 7.30% A-109 A-303 A-9 295 23.2 7.80% A-129 A-129_A A-6 295 1.8 0.60% A-PC1 A-PC3 A-10 195 9.2 4.70% A-PC1 A-PC2 A-13 145 1.7 1.20% A-PC3 A-PC3_B A-11 145 3.6 2.50% Integrating PV Solar and Wind generation with the TEC electric system 16

From Bus To Bus Branch/Xfmr Name Rated Amps Load Amperes % A-PC3_B A-PC3_C A-12 195 1.8 0.90% AC-101 AC-103 AC-4 295 39.2 13.30% AC-102 AC-102_A AC-3 295 0 0.00% AC-103 AC-126 AC-6 295 24.5 8.30% AC-103 AC-102 AC-5 230 5.5 2.40% AC-112 AC-112_B AC-13 295 12 4.10% AC-112_A AC-112 AC-12 295 13.8 4.70% AC-112_B AC-130_A AC-14 230 12 5.20% AC-118 AC-132_A AC-18 230 3.6 1.60% AC-126 AC-135_A AC-7 230 1.7 0.70% AC-126 AC-112_A AC-11 230 13.7 5.90% AC-130 AC-130_B AC-16 295 9.1 3.10% AC-130_A AC-130 AC-15 295 12 4.10% AC-130_B AC-118 AC-17 230 9.1 3.90% AC-132 AC-132_B AC-16B 230 0.1 0.00% AC-132_A AC-132 AC-19 230 3.6 1.60% AC-135 AC-135_B AC-9 295 0 0.00% AC-135_A AC-135 AC-8 295 1.8 0.60% BUS-1 EC-119 EC-1 295 14.9 5.10% BUS-1 AC-101 AC-1 295 43.2 14.70% BUS-2 BUS-1 PS BUS TIE 541 23.2 4.30% BUS-2 CL-128 CL-1 295 49 16.60% BUS-2 WC-110 WC-1 295 45.3 15.30% BUS-2 A-107 A-1 195 11.4 5.90% BUS-2 S-105 S-1 295 68.6 23.20% CL-128 CL-400_A CL-3 295 47.3 16.00% CL-128 CL-128_A CL-2 195 1.8 0.90% CL-205 CL-213 CL-40 295 7.4 2.50% CL-205 CL-215 CL-39 189 4.7 2.50% CL-206 CL-220_A CL-32 195 25 12.80% CL-207 CL-208_A CL-22 195 33.2 17.00% Integrating PV Solar and Wind generation with the TEC electric system 17

From Bus To Bus Branch/Xfmr Name Rated Amps Load Amperes % CL-208 CL-208_B CL-24 295 31.4 10.70% CL-208_A CL-208 CL-23 295 33.3 11.30% CL-208_B CL-219_A CL-25 195 31.4 16.10% CL-209 CL-214_A CL-18 195 38.2 19.60% CL-210 CL-221_A CL-14 195 29.9 15.30% CL-210 CL-313 CL-7 195 13.6 7.00% CL-211 CL-211_B CL-30 295 27.9 9.40% CL-211_A CL-211 CL-29 295 29.7 10.10% CL-211_B CL-206 CL-31 195 27.9 14.30% CL-213 CL-212 CL-41 295 3.7 1.20% CL-214 CL-214_B CL-20 295 36.6 12.40% CL-214_A CL-214 CL-19 295 38.5 13.00% CL-214_B CL-207 CL-21 295 36.6 12.40% CL-218 CL-218_B CL-37 230 17.7 7.70% CL-218_A CL-218 CL-36 230 19.6 8.50% CL-218_B CL-205 CL-38 195 17.7 9.10% CL-219 CL-219_B CL-27 295 29.7 10.10% CL-219_A CL-219 CL-26 295 31.6 10.70% CL-219_B CL-211_A CL-28 195 29.7 15.20% CL-220 CL-220_B CL-34 295 19.4 6.60% CL-220_A CL-220 CL-33 295 25.1 8.50% CL-220_B CL-218_A CL-35 195 19.4 10.00% CL-221_A CL-221 CL-15 295 30 10.20% CL-221_B CL-221 CL-16 295 41.9 14.20% CL-221_B CL-209 CL-17 195 41.9 21.50% CL-312 CL-312_C CL-12 295 0 0.00% CL-312 CL-312_A CL-9 295 12.1 4.10% CL-312_B CL-311 CL-11 195 2.8 1.40% CL-312_B CL-312 CL-10 295 2.8 1.00% CL-313 CL-312_A CL-8 195 11.9 6.10% CL-400_A CL-400 CL-4 195 48.8 25.00% Integrating PV Solar and Wind generation with the TEC electric system 18

From Bus To Bus Branch/Xfmr Name Rated Amps Load Amperes % CL-400_B CL-400 CL-5 195 43.3 22.20% CL-400_B CL-210 CL-6 295 43.3 14.70% EC-119 EC-120 EC-2 295 14.3 4.80% EC-120 EC-131 EC-3 195 5.5 2.80% EC-120 EC-121 EC-4 295 5.6 1.90% EC-121 EC-133_A EC-5 195 2.5 1.30% EC-122 EC-123_A EC-9 195 6.1 3.10% EC-123 EC-123_B EC-114 295 11 3.70% EC-123_A EC-123 EC-10 295 6.2 2.10% EC-123_B EC-124 EC-12 195 11 5.70% EC-124 EC-134_A EC-13 195 12.8 6.50% EC-133 EC-133_B EC-7 295 3.4 1.10% EC-133_A EC-133 EC-6 295 2.8 0.90% EC-133_B EC-122 EC-8 195 3.5 1.80% EC-134 EC-134_B EC-15 295 12.2 4.10% EC-134_A EC-134 EC-14 295 13.8 4.70% EC-134_B EC-202 EC-16 195 12.2 6.30% EC-202 EC-216_A EC-17 195 8.8 4.50% EC-203 EC-204 EC-24 295 2.8 0.90% EC-216 EC-216_B EC-19 295 7.1 2.40% EC-216_A EC-216 EC-18 295 8.8 3.00% EC-216_B EC-217_A EC-20 195 7.1 3.60% EC-217 EC-217_B EC-22 295 5.4 1.80% EC-217_A EC-217 EC-21 295 7.1 2.40% EC-217_B EC-203 EC-23 195 5.6 2.90% S-105 S-125 S-2 195 72.7 37.30% S-111 S-111_B S-17 295 0 0.00% S-111_A S-111 S-16 295 3.6 1.20% S-113 S-116_A S-12 230 8.8 3.80% S-114 S-117_A S-8 230 90.3 39.30% S-115 S-115_B S-6 230 87 37.80% Integrating PV Solar and Wind generation with the TEC electric system 19

From Bus To Bus Branch/Xfmr Name Rated Amps Load Amperes % S-115_A S-115 S-5 230 81.9 35.60% S-115_B S-114 S-7 230 87 37.80% S-116 S-116_B S-14 295 3.5 1.20% S-116_A S-116 S-13 295 8.8 3.00% S-116_B S-111_A S-15 230 3.5 1.50% S-117 S-117_B S-10 295 12.2 4.10% S-117_A S-117 S-9 295 90.3 30.60% S-117_B S-113 S-11 230 12.2 5.30% S-125 S-127 S-3 195 5.4 2.80% S-125 S-115_A S-4 230 81.9 35.60% WC-110 WC-302_A WC-3 230 43.3 18.80% WC-110 WC-301 WC-2 195 0.2 0.10% WC-302 WC-302_A WC-4 295 43.5 14.80% WC-302 WC-315 WC-5 295 39.8 13.50% WC-304 WC-305_A WC-9 195 31.2 16.00% WC-304 WC-318 WC-7 295 7 2.40% WC-305 WC-305_B WC-11 295 25.7 8.70% WC-305_A WC-305 WC-10 295 31.2 10.60% WC-305_B WC-306 WC-12 195 25.8 13.20% WC-306 WC-307 WC-13 195 21.2 10.90% WC-307 WC-308 WC-15 195 15.9 8.20% WC-307 WC-316 WC-14 295 1.7 0.60% WC-308 WC-309 WC-16 195 11.4 5.90% WC-309 WC-321_A WC-17 295 3.4 1.10% WC-309 WC-314_A WC-21 195 4.6 2.40% WC-314 WC-314_B WC-23 230 2.9 1.20% WC-314_A WC-314 WC-22 230 4.7 2.10% WC-314_B WC-310 WC-24 195 2.9 1.50% WC-315 WC-304 WC-6 295 38.1 12.90% WC-318 WC-319 WC-8 295 5.5 1.90% WC-321 WC-321_B WC-19 295 1.7 0.60% Integrating PV Solar and Wind generation with the TEC electric system 20

From Bus To Bus Branch/Xfmr Name Rated Amps Load Amperes % WC-321_A WC-321 WC-18 295 3.4 1.20% WC-321_B WC-317 WC-20 295 1.7 0.60% Transformer BUS-1 GS-1 TR-1 131 43 32.60% GS-2 BUS-1 TR-2 131 0 0.00% GS-4 BUS-1 TR-3 39 0 0.00% GS-5 BUS-1 TR-4 39 0 0.00% NZ-GS BUS-1 TR-7 131 0 0.00% 2.6 Steady-state conclusions In both load condition (night time off peak and day time peak load) system steady state performance (voltages and loadings) of the TAU system appears to be acceptable with the above renewable generation assumptions added in the EasyPower simulations. Dynamic simulation of the network is another factor that is required to assess system performance. Dynamic events can impose lower limits on the amount of renewable generation than the steady state simulations done above. The following section shows the results of the dynamic simulation. Integrating PV Solar and Wind generation with the TEC electric system 21

3. Dynamic simulation results NOTE: The inverter (located on each PV & Wind generator) will provide nearly no power output during the low voltage conditions on a feeder that occur during fault clearing events on the grid. Depending on the type of dynamic model used for the inverter during the simulations, the behavior of the inverters right after fault clearing shows different results. While the actual and modeled inverter results may be different, the main objective of the following analysis is to determine how the diesel generators respond to critical power system events with and without the RE inverters added to the system. 3.1 Dynamic issues with island power systems There are two key technical issues with solar and wind as renewable energy sources: 1. The random variability of the output and 2. The use of DC/AC inverters. Varying wind speeds and cloud cover directly affect the output of wind and solar generation. These variations are largely unpredictable. The power system must be able to handle the resulting output changes. The generation system must have enough reserves to adjust for the changing wind and solar output. The generation system must also be able to change quickly enough to control the system frequency as the wind and solar generation changes. This affect is called ramp-rate the speed of change in generation usually measured in KW/minute. 3.1.1 Solar and wind generation inverter voltage and frequency ridethrough All PV generators and large wind generators (>1 MW) produce direct current (DC) electricity. 11 This DC output must be converted to alternating current (AC) to be used by the electric power system. This conversion is made by an electronic DC/AC inverter. These inverters provide power to the system by following the system frequency and voltage that must be provided by conventional generation. This means that there must always be some conventional generation operating; and this, in turn, means that no system can be 100% supplied by inverter-based wind or PV generation. There is one other relevant issue with the DC/AC inverters they disconnect during unusual voltage or frequency conditions. These inverters are not usually controlled by the system operating center and use local information to operate. These inverters are designed to disconnect from the system whenever there 11. Smaller wind generators (<1 MW) produce alternating current (AC) electricity and can connect directly into the system without inverters. Integrating PV Solar and Wind generation with the TEC electric system 22

Voltage DNV KEMA Energy & Sustainability is a short circuit or other fault. They decide that a fault exists from the system voltage and frequency. Low frequencies or low voltages occur whenever there is a short circuit or fault. The settings for both these characteristics the drop out frequency and voltage can be adjusted for the inverters. These capabilities are called low-voltage ride-through and frequency ride-through. When a generator trips off line, the frequency will drop. If the frequency drops low enough for more than a second or so, solar systems like those on Rarotonga may trip offline. The effect of these solar generation trips is to push frequency down even further. The diesel generators will act to restore the frequency to the normal range. The standard ride-through limits used with solar and wind generation are shown in Figure 3 and Figure 4. 12 (These figures use a logarithmic horizontal scale to show details of the system performance.) Additional details of the voltage settings are shown in Table 11. The figures show the upper and lower limits of acceptable operation in the seconds following the initial disturbance. In both cases the acceptable range becomes narrower as times passes following the initiating event. Figure 3: Standard voltage ride-through limits 1.20 Upper limit 1.00 0.80 Acceptable region 0.60 0.40 Lower limit 0.20 0.00 0.01 0.1 1 10 100 Seconds 12. There are a range of voltage and frequency standards applied worldwide a survey can be found in a report by TransPower New Zealand, Generator Fault Ride Through (FRT) Investigation, Stage 1 Literature Review, Feb 2009. Integrating PV Solar and Wind generation with the TEC electric system 23

Table 11: Standard voltage ride-through settings Low voltage set point High voltage set point Voltage (pu) Voltage (pu) 0.00 0.02 0.00 0.00 0.20 1.200 0.02 0.20 0.45 0.20 0.50 1.175 0.20 2.00 0.65 0.50 1.00 1.150 2.00 10.00 0.90 1.00 10.00 1.100 Figure 4: Standard frequency ride-through limits Frequency (Hz) 52.5 52.0 51.5 51.0 50.5 Acceptable region 50.0 49.5 49.0 48.5 48.0 47.5 0.1 1 10 100 1000 10000 Seconds 3.1.2 Island small system sizes The nature of island electric power systems presents several challenges related to integrating renewable energy resources such as wind and solar. Most of these challenges are related to the small overall system size and the fact that there are no interconnections to other systems (as would often be the case in mainland system). The small electrical size usually means that there are only a few generating units usually diesel powered operating at any time. The loss of one of these generators can often result in fairly wide swings in system voltage and frequency. Integrating PV Solar and Wind generation with the TEC electric system 24

3.1.3 Technical solutions Integrating solar and wind renewable generation into island systems can be managed. System resources must be continuously managed to balance load and generation as load and renewable generation fluctuate. 13 The key limitations are ramp rates, and low-voltage and frequency ride-through settings of inverters. One approach is to limit the total amount of renewable generation that can be installed and/or limit the maximum size of any one facility. These limits mitigate ramp-rate and ride-through issues. They can also be mitigated with storage batteries that vary power and energy for a few minutes to mitigate the outage of a renewable generator. Such battery systems can also mitigated large changes in output from varying wind or solar radiation. The battery would provide watts when frequency is low. This is not a normal operating mode for backup batteries, so it is important that any batteries used this way have this capability and are set accordingly. 3.2 Base case daytime peak load (baseline without RE) In order to establish a baseline or point of reference for dynamic frequency and voltage response on Rarotonga, a selected set of disturbances was run without any additional RE generation. Below are results from a disturbance on the pre-re base case (five diesel generators running with daytime peak load condition and no renewable generation). 13 EPC is planning to install a new Avatiu Valley Power Station controller in the coming year at the diesel Avatiu Valley Power Station on Rarotonga. Data for this new controller was unavailable to KEMA and has not been modeled in the analysis discussed in this report. Integrating PV Solar and Wind generation with the TEC electric system 25

Bus 1 Voltage (pu) DNV KEMA Energy & Sustainability Disturbance: Loss of largest diesel unit (GS-7); dispatched at 2000 kw in the base case. Figure 5: Trip of GS-7 (No fault); Frequency Response Base Case - loss of largest unit (2000 kw) Bus 1 Frequency (Hz) 50.1 50 49.9 49.8 49.7 49.6 49.5 49.4 49.3 49.2 49.1 0 1 2 3 4 5 6 7 8 9 10 Figure 6: Trip of GS-7 (No fault); Voltage Response 1.02 Base Case - loss of largest unit (2000 kw) 1.01 1 0.99 0.98 0.97 0.96 0 1 2 3 4 5 6 7 8 9 10 As Figure 5 shows, the frequency drops to 49.4 Hz but recovers relatively quickly. Figure 6 shows the voltage during the same conditions. The voltage varies from 0.98 pu to 1.015 pu. The system demonstrates stability and a solid 1.0 pu bus voltage at the Avatiu Valley Power Station post contingency. Integrating PV Solar and Wind generation with the TEC electric system 26

Disturbance: The following figures reflect system response for an un-faulted trip of the Cross Line 11 kv Feeder, which is the feeder with the heaviest demand on Rarotonga in the base case model. Figure 7: Trip of Cross Line Feeder (No fault); Frequency Response Base Case - loss of Cross Line Feeder Bus 1 Frequency (Hz) Bus 1 Voltage (pu) 50.3 50.25 50.2 50.15 50.1 50.05 50 49.95 49.9 49.85 0 1 2 3 4 5 6 7 8 9 10 Figure 8: Trip of Cross Line Feeder (No fault); Voltage Response 1.035 Base Case - loss of Cross Line Feeder 1.03 1.025 1.02 1.015 1.01 1.005 0 1 2 3 4 5 6 7 8 9 10 The frequency rises as high as 50.25 Hz but the power system remains stable and is returning to normal 50 Hz in an acceptable amount of time. Loss of the Cross Line feeder causes the per unit voltage at the Integrating PV Solar and Wind generation with the TEC electric system 27

power plant to increase from 1.015 to 1.03. This voltage fluctuation however remains within acceptable post-contingent voltage limits. Disturbance: Apply a 3-phase fault to the Cross Line 11 kv Feeder at first bus outside of Avatiu Valley Power Station at Time 1 second, then clear the fault by opening the feeder breaker six cycles later. Figure 9: Trip Line Feeder with Close-in Fault; Frequency Response Base Case - Cross Line Feeder fault Bus 1 Frequency (Hz) Bus 1 Voltage (pu) 50.5 50.4 50.3 50.2 50.1 50 49.9 49.8 49.7 49.6 0 1 2 3 4 5 6 7 8 9 Figure 10: Trip Cross Line Feeder with Close-in Fault; Voltage Response 1.2 Base Case - Cross Line Feeder fault 1 0.8 0.6 0.4 0.2 0 0 1 2 3 4 5 6 7 8 9 Integrating PV Solar and Wind generation with the TEC electric system 28

Bus 1 Frequency (Hz) DNV KEMA Energy & Sustainability The two figures above show the frequency and voltage response at the Avatiu Valley Power Station bus for a fault and subsequent clearing on the Cross Line feeder. The frequency initially dips during the fault then increases as the faulted feeder breaker is opened and load is removed and eventually returns to 50 Hz. The voltage response shows the depressed voltage during the fault condition and then voltage recovery to slightly above 1.0 pu after clearing the fault. 3.3 Base case off-peak load (baseline without RE) Similar to section 3.2, a selected set of disturbances was run without any additional RE generation in model representing off-peak or night time loading conditions. Below are results from several disturbances on the pre-re base case (three diesel generators running with nighttime off-peak load condition and no renewable generation). Disturbance: Loss of largest diesel unit (GS-7); dispatched at 1000 kw in the base case. Figure 11: Trip of GS-7 (No fault); Frequency Response 50.2 Base Case Light Load - loss of largest unit (1000 kw) 50 49.8 49.6 49.4 49.2 0 1 2 3 4 5 6 7 8 9 10 Integrating PV Solar and Wind generation with the TEC electric system 29

Bus 1 Frequency (Hz) DNV KEMA Energy & Sustainability Figure 12: Trip of GS-7 (No fault); Voltage Response Bus 1 Voltage (pu) 1.02 1.015 1.01 1.005 1 0.995 0.99 0.985 0.98 Base Case Light Load - loss of largest unit (1000 kw) 0 1 2 3 4 5 6 7 8 9 10 As Figure 11shows, the frequency drops to 49.5 Hz and recovers relatively quickly. Figure 6 shows the voltage during the same conditions. The voltage varies from 0.995 pu to 1.015 pu and returns to a stable voltage level of nearly 1.01 pu post contingency. Disturbance: The following figures reflect system response for an un-faulted trip of the Cross Line 11 kv Feeder, which is the feeder with the heaviest demand on Rarotonga in the base case model. Figure 13: Trip of Cross Line Feeder (No fault); Frequency Response Base Case Light Load - loss of Cross Line Feeder 50.25 50.2 50.15 50.1 50.05 50 49.95 49.9 0 1 2 3 4 5 6 7 8 9 10 Integrating PV Solar and Wind generation with the TEC electric system 30

Figure 14: Trip of Cross Line Feeder (No fault); Voltage Response Bus 1 Voltage (pu) 1.022 1.02 1.018 1.016 1.014 1.012 1.01 Base Case Light Load - loss of Cross Line Feeder 0 1 2 3 4 5 6 7 8 9 10 The frequency rises as high as 50.2 Hz but the power system remains stable and is returning to normal 50 Hz in an acceptable amount of time. Loss of the Cross Line feeder causes the per unit voltage at the power plant to increase from 1.015 as high as 1.021. This voltage fluctuation however remains within acceptable post-contingent voltage limits and the bus voltage returns to the same 1.015 once the transient response to the disturbance has passed. Integrating PV Solar and Wind generation with the TEC electric system 31

Bus 1 Voltage (pu) DNV KEMA Energy & Sustainability Disturbance: Apply a 3-phase fault to the Cross Line 11 kv Feeder at first bus outside of Avatiu Valley Power Station at Time 1 second, then clear the fault by opening the feeder breaker six cycles later. Figure 15: Trip Line Feeder with Close-in Fault; Frequency Response Base Case Light Load - Cross Line Feeder fault Bus 1 Frequency (Hz) 50.4 50.3 50.2 50.1 50 49.9 49.8 49.7 0 1 2 3 4 5 6 7 8 9 Figure 16: Trip Cross Line Feeder with Close-in Fault; Voltage Response 1.2 1 0.8 0.6 0.4 0.2 0 Base Case Light Load - Cross Line Feeder fault 0 1 2 3 4 5 6 7 8 9 The two figures above show the frequency and voltage response at the Avatiu Valley Power Station bus for a fault and subsequent clearing on the Cross Line feeder. The frequency initially dips during the fault Integrating PV Solar and Wind generation with the TEC electric system 32

then increases as the faulted feeder breaker is opened and load is removed and eventually returns to 50 Hz. The voltage response shows the depressed voltage during the fault condition and then voltage recovery to 1.0 pu or higher after clearing the fault. 3.4 Night case with wind generation Disturbance: Trip of largest diesel unit (GS-7) without fault; dispatched at 600 kw, in the night time offpeak case with the wind generation output at 1,300 kw. Figure 17: Off-Peak with Wind; Loss of GS-7 (No fault); Frequency Response Off-Peak Load with Wind Loss of largest unit (600 kw) Bus 1 Frequency (Hz) 50.2 50 49.8 49.6 49.4 49.2 49 0 1 2 3 4 5 6 7 8 9 10 Integrating PV Solar and Wind generation with the TEC electric system 33

Figure 18: Off-Peak with Wind; Loss of GS-7 (No fault); Voltage Response Bus 1 Voltage (pu) 1.02 1.015 1.01 1.005 1 0.995 0.99 0.985 Off-Peak Load with Wind Loss of largest unit (600 kw) 0 1 2 3 4 5 6 7 8 9 10 Figure 17 above shows the frequency response in the off-peak, light load model with 1300 kw of wind generation resources and the loss of GS-7 which is dispatched at 600 kw in the model. The frequency dips to 49.4 Hz and returns to 50 Hz. Figure 18 shows the voltage response for the same contingency condition. Voltage at the Avatiu Valley Power Station bus decreases as expected but remains stable and well within the acceptable post-contingent voltage range. Integrating PV Solar and Wind generation with the TEC electric system 34

Bus 1 Voltage (pu) DNV KEMA Energy & Sustainability Disturbance: Apply a 3-phase fault to the East Coast 11 kv Feeder at first bus outside of Avatiu Valley Power Station then clear the fault by opening the feeder breaker. Figure 19: East Coast Feeder fault; Frequency Response Off-Peak Load with Wind East Coast Feeder fault Bus 1 Frequency (Hz) 50.1 50 49.9 49.8 49.7 49.6 49.5 0 1 2 3 4 5 6 7 8 9 Figure 20: East Coast Feeder fault; Voltage Response 1.2 Off-Peak Load with Wind East Coast Feeder fault 1 0.8 0.6 0.4 0.2 0 0 1 2 3 4 5 6 7 8 9 Figure 19 above shows the frequency response for a close-in fault on the East Coast Feeder and subsequent clearing of the fault. The East Coast feeder was chosen for this scenario because the largest wind generation resource in the model at Metavera is interconnected to this feeder. The transient Integrating PV Solar and Wind generation with the TEC electric system 35

Bus Frequency (Hz) DNV KEMA Energy & Sustainability frequency response swings from 49.75 Hz to approximately 50.05 Hz but remains stable and does not dip below 49.7 Hz. Figure 18 shows the voltage response for the same contingency condition. Voltage at the Avatiu Valley Power Station bus is depressed to near zero during the fault, but recovers quickly and returns to 1.0 pu. Disturbance: Loss of largest wind generating unit (Metevera); dispatched at 900 kw in the night time offpeak case. Figure 21: Loss of Largest Wind Resource (No fault); Frequency Response Off-Peak Load with Wind Loss of Metevera Wind Generation 50.1 50 49.9 49.8 49.7 49.6 49.5 49.4 49.3 0 1 2 3 4 5 6 7 8 9 10 Integrating PV Solar and Wind generation with the TEC electric system 36

Figure 22: Loss of Largest Wind Resource (No fault); Voltage Response 1.04 Off-Peak Load with Wind Loss of Metevera Wind Generation Bus Voltage (pu) 1.03 1.02 1.01 1 0.99 0.98 0 1 2 3 4 5 6 7 8 9 10 Figure 21 above shows the frequency response in the off-peak, light load model for loss of the Metavera wind generation site which is dispatched at 900 kw in the model. The frequency at the customer transformer closest to the Metevera wind farm dips to 49.6 Hz but is returning to 50 Hz. Figure 22 shows the voltage response for the same contingency condition at the nearest customer transformer bus. Voltage decreases as expected but remains stable and well within the acceptable post-contingent voltage range. 3.5 Day case with wind and PV generation Disturbance: Loss of largest diesel unit (GS-7); dispatched at 1,000 kw in the day time peak case. Integrating PV Solar and Wind generation with the TEC electric system 37

Bus 1 Voltage (pu) DNV KEMA Energy & Sustainability Figure 23: Peak with Wind & PV; Loss of GS-7 (No fault); Frequency Response Peak Load with Wind and PV Loss of largest unit (1000 kw) Bus 1 Frequency (Hz) 50.1 50 49.9 49.8 49.7 49.6 49.5 49.4 49.3 49.2 0 1 2 3 4 5 6 7 8 9 10 Figure 24: Peak with Wind & PV; Loss of GS-7; Voltage Response 1.02 1 0.98 0.96 0.94 0.92 0.9 0.88 0.86 Peak Load with Wind and PV Loss of largest unit (1000 kw) 0 1 2 3 4 5 6 7 8 9 10 Figure 23 above shows the frequency response in the peak load model with 1300 kw of wind generation resources and 2000 kw of PV resources and the loss of GS-7 which is dispatched at 1000 kw. The frequency slowly drops to 49.5 Hz but recovers. Figure 24 shows the voltage response for the same contingency condition. Voltage at the Avatiu Valley Power Station bus decreases to 0.92 pu but returns to 0.96 pu which is within the acceptable range for post-contingent voltages. Integrating PV Solar and Wind generation with the TEC electric system 38

Bus 1 Voltage (pu) DNV KEMA Energy & Sustainability Disturbance: Apply a 3-phase fault to the Seaport 11 kv Feeder at first bus outside of Avatiu Valley Power Station then clear the fault by opening the feeder breaker. Figure 25: Seaport Feeder fault; Frequency Response Peak Load with Wind and PV Seaport Feeder fault Bus 1 Frequency (Hz) 50.2 50 49.8 49.6 49.4 49.2 49 0 1 2 3 4 5 6 7 8 9 Figure 26: Seaport Feeder fault; Voltage Response 1.2 Peak Load with Wind and PV Seaport Feeder fault 1 0.8 0.6 0.4 0.2 0 0 1 2 3 4 5 6 7 8 9 Figure 25 above shows the frequency response for a close-in fault on the Seaport Feeder and subsequent clearing of the fault. The Seaport feeder was chosen for this scenario because the Sun City PV installation is interconnected to this feeder. The transient frequency response swings from 49.4 Hz to Integrating PV Solar and Wind generation with the TEC electric system 39

Bus Voltage (pu) DNV KEMA Energy & Sustainability 50 Hz and remains stable. Figure 26 shows the voltage response for the same contingency condition. Voltage at the Avatiu Valley Power Station bus is depressed to near zero during the fault, but recovers quickly and returns to 1.0 pu. Disturbance: Loss of Sun City PV Installation; dispatched at 2000 kw in the day time peak model. Figure 27: Loss of Sun City 2000 kw PV (No fault); Frequency Response Peak Load with Wind and PV Loss of Sun City PV Bus Frequency (Hz) 50.2 50 49.8 49.6 49.4 49.2 49 48.8 48.6 0 1 2 3 4 5 6 7 8 9 10 Figure 28: Loss of Sun City 2000 kw PV (No fault); Voltage Response 1.04 Peak Load with Wind and PV Loss of Sun City PV 1.02 1 0.98 0.96 0.94 0.92 0 1 2 3 4 5 6 7 8 9 10 Integrating PV Solar and Wind generation with the TEC electric system 40

Figure 27 above shows the frequency response at the customer transformer closest to the interconnection with Sun City in the on-peak load model for loss of the 2,000 kw Sun City generation input. The frequency at the point of interconnection dips to 49.2 Hz but recovers relatively quickly. Figure 28 shows the voltage response for the same contingency condition. Voltage at the PV interconnection bus decreases to 0.96 pu but recovers and remains stable at 0.98 pu. 3.6 Equipment controls With the added wind and PV generation require reviewing the control devices on the electric power network. The inverter controls on the proposed wind and PV generation must be able to ride thru when the network exhibits either low voltage and/or low frequency. These settings must be coordinated between the inverters settings and the protective devices that are also located on the electrical network. The analysis of the type and settings of the protective devices are outside the scope of this project and will only be determined once the inverters are selected. Integrating PV Solar and Wind generation with the TEC electric system 41

4. Revised Modeling and Scenarios Once the above scenarios were presented to TAU, additional sensitivities were requested to examine different configurations of distributed renewable resources based on updated wind data and a report on potential sites for large-scale PV installations. 14 Three additional scenarios were modeled, two using the daytime peak model with an increased penetration of PV. The third scenario uses the night time off-peak base case and a single large-scale wind farm (4000 kw) at Raemaru. 4.1 Revised Day case with wind and PV generation In this scenario, the day time base case was revised to include updated capacities for the Metevera and West Wigmores wind generation sites. Their output was reduced to 100 kw each for a total reduction in wind generation of 600 kw. Additional PV sites were modeled based on the priority ranking shown in Table 12 below. PV installation rankings 1 through 4 were modeled in the day-time peak case with wind and PV. Table 12: Potential Large-scale PV Installations 15 14 Source: Potential Sites for Grid Assessment J000898 Rarotonga Large Scale PV; Southern Perspectives; 12/12/2012 15 Source: Potential Sites for Grid Assessment J000898 Rarotonga Large Scale PV; Southern Perspectives; 12/12/2012 Integrating PV Solar and Wind generation with the TEC electric system 42

The following are the assumptions used in the power flow model for the Project Scenario described in the following section: Model from first round steady-state and dynamic stability analysis Day-time Peak Case with Wind and PV. Load is 4,830 kw on-peak (DAY) load. Two diesel generators were assumed to be running; 2 units online dispatched at 234 kw and 250 kw, respectively. At least one diesel unit is needed to provide a regulated voltage and frequency source for the island. Running two diesel units allows either diesel unit to continue regulating voltage and frequency if the other unit trips off line. This is typically referred to as an n-1 resource criterion. Six PV sites were modeled for a total of 3778 kw as shown below in Table 13. Three wind generation sites as shown in Table 13; these are the same wind generation sites used in the first round of analysis but with revised capacities for Matavera and West Wigmores sites per an e-mail from Shay Brazier. Integrating PV Solar and Wind generation with the TEC electric system 43

Table 13: Renewable Resources Summary Site DRG Type Capacity (kw) Assumed Output (kw) FDR Matevera Wind 100 100 EAST COAST Hospital Wind 500 500 AIRPORT West Wigmores Wind 100 100 CROSS LINE Sun City PV 2000 2000 SEAPORT TSA Stadium PV 195 195 AIRPORT Telecom PV 210 210 CROSS LINE Cultural Center PV 155 155 EAST COAST Airport Hanger PV 120 120 AIRPORT Airport Terminal PV 98 98 AIRPORT Panama Dump PV 1,300 1,000 SEAPORT Table 14 below shows the generation and load summary for the revised model and Table 15 shows the base case dispatch conditions modeled for the PV projects, wind turbines and diesel generators. Table 14: Day time peak case with revised RE system summary Total MW Mvar MVA pf Generation in System 4.962 1.575 5.206 0.9531 Load in System 4.830 2.343 5.368 0.900 Shunt Load in System 0 0 Losses in System 0.165-0.768 Check of Balance 0 0 Integrating PV Solar and Wind generation with the TEC electric system 44

Table 15: Day time peak case with revised RE generation summary Name Generator Type Solution Rated MVA MW Mvar MVA PF Vpu Deg GS-1 Sw 2.000 0.234 0.779 0.813 0.288 1.020 0 GS-7 PQG 2.700 0.250 0.150 0.292 0.857 0.995-0.22 INV_CLTR_PV PQG 0.200 0.155 0.022 0.157 0.990 0.992.13 INV_DUMP_PV PQG 1.500 1.000 0.142 1.010 0.990 1.023 0.53 INV_PVC_1 PQG 1.500 1.000 0.142 1.010 0.990 1.027 0.64 INV_PCV_2 PQG 1.500 1.000 0.142 1.010 0.990 1.027 0.64 INV_TLCM_PV PQG 0.250 0.210 0.032 0.212 0.989 0.996 0.29 INV_TRMHNG_PV PQG 0.250 0.218 0.033 0.220 0.989 1.001 0.41 INV_TSA_PV PQG 0.250 0.195 0.028 0.197 0.990 1.002 0.37 INV_WND_HOS PQG 1.500 0.500 0.070 0.505 0.990 1.002 0.17 INV_WND_MET PQG 0.150 0.100 0.020 0.102 0.981 0.988 0.12 INV_WND_WIG PQG 0.150 0.100 0.015 0.101 0.989 0.978 0.14 In this revised day time case, a total of 700 kw of wind generation and 3,778 kw of PV generation were modeled taking the conventional generation output down to 484 kw. With this level of wind and PV penetration the system experienced no feeder or transformer overloads as shown in Table 16. Table 16: Day time peak case with revised RE feeder and transformer loading summary From Bus To Bus Branch/Xfmr Name Feeder Rated Amps Load Amperes % A-107 A-108 A-2 295 18.2 6.20% A-107 A-PC1 A-3 195 6 3.10% A-108 A-108_B A-7 295 16.3 5.50% A-108 A-108_A A-4 195 6.9 3.50% Integrating PV Solar and Wind generation with the TEC electric system 45

From Bus To Bus Branch/Xfmr Name Rated Amps Load Amperes % A-108 A-129_A A-5 195 1.7 0.90% A-108_B A-109 A-8 295 21.5 7.30% A-109 A-303 A-9 295 23.1 7.80% A-129 A-129_A A-6 295 1.8 0.60% A-PC1 A-PC2 A-13 145 1.7 1.20% A-PC1 A-PC3 A-10 195 9.2 4.70% A-PC3 A-PC3_B A-11 145 3.6 2.50% A-PC3_B A-PC3_C A-12 195 1.8 0.90% AC-101 AC-103 AC-4 295 28.7 9.70% AC-102 AC-102_A AC-3 295 0 0.00% AC-103 AC-126 AC-6 295 24.7 8.40% AC-103 AC-102 AC-5 230 6.1 2.70% AC-112 AC-112_B AC-13 295 12 4.10% AC-112_A AC-112 AC-12 295 13.8 4.70% AC-112_B AC-130_A AC-14 230 12 5.20% AC-118 AC-132_A AC-18 230 3.6 1.60% AC-126 AC-112_A AC-11 230 13.7 6.00% AC-126 AC-135_A AC-7 230 1.7 0.70% AC-130 AC-130_B AC-16 295 9.1 3.10% AC-130_A AC-130 AC-15 295 12 4.10% AC-130_B AC-118 AC-17 230 9.1 4.00% AC-132 AC-132_B AC-16B 230 0.1 0.00% AC-132_A AC-132 AC-19 230 3.6 1.60% AC-135 AC-135_B AC-9 295 0 0.00% AC-135_A AC-135 AC-8 295 1.8 0.60% BUS-1 EC-119 EC-1 295 28.1 9.50% Integrating PV Solar and Wind generation with the TEC electric system 46

From Bus To Bus Branch/Xfmr Name Rated Amps Load Amperes % BUS-1 AC-101 AC-1 295 32.6 11.10% BUS-2 CL-128 CL-1 295 60.2 20.40% BUS-2 BUS-1 PS BUS TIE 541 47.6 8.80% BUS-2 WC-110 WC-1 295 45.5 15.40% BUS-2 A-107 A-1 195 13.8 7.10% BUS-2 S-105 S-1 295 119.8 40.60% CL-128 CL-400_A CL-3 295 58.5 19.80% CL-128 CL-128_A CL-2 195 1.8 0.90% CL-205 CL-215 CL-39 189 4.8 2.50% CL-205 CL-213 CL-40 295 7.5 2.50% CL-206 CL-220_A CL-32 195 25.2 12.90% CL-207 CL-208_A CL-22 195 33.6 17.20% CL-208 CL-208_B CL-24 295 31.8 10.80% CL-208_A CL-208 CL-23 295 33.7 11.40% CL-208_B CL-219_A CL-25 195 31.8 16.30% CL-209 CL-214_A CL-18 195 38.6 19.80% CL-210 CL-313 CL-7 195 13.7 7.00% CL-210 CL-221_A CL-14 195 40.7 20.80% CL-211 CL-211_B CL-30 295 28.1 9.50% CL-211_A CL-211 CL-29 295 30 10.20% CL-211_B CL-206 CL-31 195 28.1 14.40% CL-213 CL-212 CL-41 295 3.7 1.30% CL-214 CL-214_B CL-20 295 37 12.50% CL-214_A CL-214 CL-19 295 38.9 13.20% CL-214_B CL-207 CL-21 295 37 12.50% CL-218 CL-218_B CL-37 230 17.9 7.80% Integrating PV Solar and Wind generation with the TEC electric system 47

From Bus To Bus Branch/Xfmr Name Rated Amps Load Amperes % CL-218_A CL-218 CL-36 230 19.8 8.60% CL-218_B CL-205 CL-38 195 17.9 9.20% CL-219 CL-219_B CL-27 295 30 10.20% CL-219_A CL-219 CL-26 295 31.9 10.80% CL-219_B CL-211_A CL-28 195 30 15.40% CL-220 CL-220_B CL-34 295 19.6 6.70% CL-220_A CL-220 CL-33 295 25.4 8.60% CL-220_B CL-218_A CL-35 195 19.6 10.10% CL-221_A CL-221 CL-15 295 40.7 13.80% CL-221_B CL-221 CL-16 295 42.3 14.30% CL-221_B CL-209 CL-17 195 42.3 21.70% CL-312 CL-312_C CL-12 295 0 0.00% CL-312 CL-312_A CL-9 295 12.3 4.20% CL-312_B CL-312 CL-10 295 2.8 1.00% CL-312_B CL-311 CL-11 195 2.8 1.50% CL-313 CL-312_A CL-8 195 12 6.20% CL-400_A CL-400 CL-4 195 59.8 30.70% CL-400_B CL-210 CL-6 295 54.3 18.40% CL-400_B CL-400 CL-5 195 54.3 27.80% EC-119 EC-120 EC-2 295 27.1 9.20% EC-120 EC-131 EC-3 195 3.4 1.80% EC-120 EC-121 EC-4 295 26.2 8.90% EC-121 EC-133_A EC-5 195 20.9 10.70% EC-122 EC-123_A EC-9 195 15.7 8.10% EC-123 EC-123_B EC-114 295 10.4 3.50% EC-123_A EC-123 EC-10 295 15.8 5.40% Integrating PV Solar and Wind generation with the TEC electric system 48

From Bus To Bus Branch/Xfmr Name Rated Amps Load Amperes % EC-123_B EC-124 EC-12 195 10.4 5.30% EC-124 EC-134_A EC-13 195 8.6 4.40% EC-133 EC-133_B EC-7 295 19.3 6.50% EC-133_A EC-133 EC-6 295 21 7.10% EC-133_B EC-122 EC-8 195 19.3 9.90% EC-134 EC-134_B EC-15 295 12.3 4.20% EC-134_A EC-134 EC-14 295 14.1 4.80% EC-134_B EC-202 EC-16 195 12.4 6.40% EC-202 EC-216_A EC-17 195 8.9 4.60% EC-203 EC-204 EC-24 295 2.8 1.00% EC-216 EC-216_B EC-19 295 7.2 2.40% EC-216_A EC-216 EC-18 295 9 3.00% EC-216_B EC-217_A EC-20 195 7.2 3.70% EC-217 EC-217_B EC-22 295 5.5 1.90% EC-217_A EC-217 EC-21 295 7.2 2.50% EC-217_B EC-203 EC-23 195 5.6 2.90% S-105 S-125 S-2 195 123.9 63.60% S-111 S-111_B S-17 295 0 0.00% S-111_A S-111 S-16 295 3.6 1.20% S-113 S-116_A S-12 230 8.8 3.80% S-114 S-117_A S-8 230 89.8 39.10% S-115 S-115_B S-6 230 138.3 60.10% S-115_A S-115 S-5 230 133.2 57.90% S-115_B S-114 S-7 230 138.3 60.10% S-116 S-116_B S-14 295 3.4 1.20% S-116_A S-116 S-13 295 8.8 3.00% Integrating PV Solar and Wind generation with the TEC electric system 49

From Bus To Bus Branch/Xfmr Name Rated Amps Load Amperes % S-116_B S-111_A S-15 230 3.5 1.50% S-117 S-117_B S-10 295 12.2 4.10% S-117_A S-117 S-9 295 89.8 30.40% S-117_B S-113 S-11 230 12.2 5.30% S-125 S-127 S-3 195 5.4 2.80% S-125 S-115_A S-4 230 133.2 57.90% WC-110 WC-302_A WC-3 230 43.6 18.90% WC-110 WC-301 WC-2 195 0.2 0.10% WC-302 WC-315 WC-5 295 40.1 13.60% WC-302 WC-302_A WC-4 295 43.8 14.80% WC-304 WC-305_A WC-9 195 31.4 16.10% WC-304 WC-318 WC-7 295 7.1 2.40% WC-305 WC-305_B WC-11 295 25.9 8.80% WC-305_A WC-305 WC-10 295 31.4 10.60% WC-305_B WC-306 WC-12 195 25.9 13.30% WC-306 WC-307 WC-13 195 21.3 10.90% WC-307 WC-308 WC-15 195 16 8.20% WC-307 WC-316 WC-14 295 1.7 0.60% WC-308 WC-309 WC-16 195 11.5 5.90% WC-309 WC-321_A WC-17 295 3.4 1.20% WC-309 WC-314_A WC-21 195 4.6 2.40% WC-314 WC-314_B WC-23 230 2.9 1.30% WC-314_A WC-314 WC-22 230 4.8 2.10% WC-314_B WC-310 WC-24 195 2.9 1.50% WC-315 WC-304 WC-6 295 38.4 13.00% WC-318 WC-319 WC-8 295 5.5 1.90% Integrating PV Solar and Wind generation with the TEC electric system 50

From Bus To Bus Branch/Xfmr Name Rated Amps Load Amperes % WC-321 WC-321_B WC-19 295 1.7 0.60% WC-321_A WC-321 WC-18 295 3.4 1.20% WC-321_B WC-317 WC-20 295 1.7 0.60% Transformer TR-1 BUS-1 GS-1 2500 793 31.70% TR-2 GS-2 BUS-1 2500 0 0.00% TR-3 GS-4 BUS-1 750 0 0.00% TR-4 GS-5 BUS-1 750 0 0.00% TR-7 NZ-GS BUS-1 2500 0 0.00% TX-RE-CLTR_PV EC-131 CLTR_PV_ACBUS 250 156 62.50% TX-RE-DUMP_PV S-114 DUMP_PV_ACBUS 1500 1008 67.20% TX-RE-HOSPITAL A-303 WND_HOS_AC 1500 504 33.60% TX-RE-MET EC-134_A WND_AC_MET 150 102 67.70% TX-RE-PV S-117 PVC_AC 2500 2016 80.60% TX-RE-TLCM_PV AC-102 TLCM_PV_ACBUS 250 212 84.70% TX-RE-TRMHNG_PV A-PC1 TRMHNG_PV_ACBUS 250 220 87.90% TX-RE-TSA_PV A-108_A TSA_PV_ACBUS 250 196 78.50% TX-RE-WIGMORES CL-221 WND_1_WIG 150 101 67.20% 4.1.1 Day case with revised RE Dynamic Stability Analysis Disturbance: Loss of largest diesel unit (GS-7); dispatched at 250 kw in the day time peak case. Integrating PV Solar and Wind generation with the TEC electric system 51

Bus 1 Voltage (pu) DNV KEMA Energy & Sustainability Figure 29: Peak with revised Wind & PV; Loss of GS-7 (No fault); Frequency Response Peak Load with Wind and PV Loss of largest unit (250 kw) Bus 1 Frequency (Hz) 50.05 50 49.95 49.9 49.85 49.8 49.75 49.7 0 1 2 3 4 5 6 7 8 9 10 Figure 30: Peak with revised Wind & PV; Loss of GS-7 (No fault); Voltage Response 1.01 1.005 1 0.995 0.99 0.985 0.98 0.975 0.97 Peak Load with Wind and PV Loss of largest unit (250 kw) 0 1 2 3 4 5 6 7 8 9 10 Figure 29 above shows the frequency response in the peak load model for the loss of GS-7 which is dispatched at 250 kw. Figure 30 is the voltage response at the Avatiu Valley Power Station bus for the same contingency. The reduced output of GS-7 due to the additional PV installations means a much Integrating PV Solar and Wind generation with the TEC electric system 52

smaller impact to the power system when the unit goes offline, and the remaining diesel unit is able to recover and maintain power system stability. Disturbance: Apply a 3-phase fault to the Seaport 11 kv Feeder at first bus outside of Avatiu Valley Power Station then clear the fault by opening the feeder breaker. The Seaport feeder was chosen for this scenario because both the Sun City and Panama Dump PV installations interconnect to this feeder. So a fault at this location trips the feeder the feeder with both its connected load and PV generation with a total output of 3,000 kw. Figure 31: Seaport Feeder fault; Frequency Response Peak Load with Wind and PV Seaport Feeder fault Bus 1 Frequency (Hz) 50.2 50 49.8 49.6 49.4 49.2 49 48.8 48.6 0 1 2 3 4 5 6 7 8 9 Integrating PV Solar and Wind generation with the TEC electric system 53

Figure 32: Seaport Feeder fault; Voltage Response 1.2 Peak Load with Wind and PV Seaport Feeder fault Bus 1 Voltage (pu) 1 0.8 0.6 0.4 0.2 0 0 1 2 3 4 5 6 7 8 9 Figure 31above shows the frequency response for a close-in fault on the Seaport Feeder and subsequent clearing of the fault. The transient frequency response swings below 49.1 Hz but returns to 50 Hz within an acceptable amount of time. Figure 32 shows the voltage response for the same contingency condition. Voltage at the Avatiu Valley Power Station bus is depressed to near zero during the fault, but recovers quickly and returns to 1.0 pu. The magnitude of this voltage drop could result in tripping of other RE units on Rarotonga if they do not have fault ride through capability. KEMA has assumed that the other RE units will have suitable fault ride through capability so they were not tripped for this simulation. Disturbance: Loss of Sun City PV Installation; dispatched at 2000 kw in the day time peak model. Integrating PV Solar and Wind generation with the TEC electric system 54

Bus Voltage (pu) DNV KEMA Energy & Sustainability Figure 33: Loss of Sun City 2000 kw PV (No fault); Frequency Response Peak Load with Wind and PV Loss of Sun City PV Bus Frequency (Hz) 50.2 50 49.8 49.6 49.4 49.2 49 48.8 48.6 0 1 2 3 4 5 6 7 8 9 10 Figure 34: Loss of Sun City 2000 kw PV (No fault); Voltage Response 1.04 1.03 1.02 1.01 1 0.99 0.98 0.97 0.96 Peak Load with Wind and PV Loss of Sun City PV 0 1 2 3 4 5 6 7 8 9 10 The frequency and voltage response at the customer transformer closest to the Sun City PV installation is shown in Figure 33 and Figure 34, above, for loss of the Sun City PV generation as this is the largest single PV installation modeled on the island. The frequency at the closest customer bus dips to 49.1 Hz but recovers relatively quickly. The voltage at this bus starts at 1.03 pu and only drops to 0.99 pu before returning to 1.0 pu. Integrating PV Solar and Wind generation with the TEC electric system 55

4.2 Revised Night case with wind generation In this scenario, the night time case was revised to include a large-scale wind installation at the Raemaru site, interconnected to the West Coast 11 kv feeder. The original wind generation was removed from the model, so the sole wind generation online is at the Raemaru facility. The following are the assumptions used in the power flow model for the Project Scenario described in the following section: Model from first round steady-state and dynamic stability analysis. Load is 2,400 kw off-peak (NIGHT) load. Two diesel generators were assumed to be running; 2 units online dispatched at 246 kw and 250 kw, respectively. At least one diesel unit is needed to provide a regulated voltage and frequency source for the island. Running two diesel units allows either diesel unit to continue regulating voltage and frequency if the other unit trips off line. This is typically referred to as an n-1 resource criterion. Wind generation at Hospital, West Wigmores and Matevera sites removed. Wind generation site at Raemaru added to the night-time off-peak case; dispatched at 2000 kw. Table 17 below shows the generation and load summary for the revised model and Table 18 shows the base case dispatch conditions modeled for the PV project, wind turbines and diesel generators. Table 17: Night time off-peak case with revised RE system summary Total MW Mvar MVA pf Generation in System 2.496 0.315 2.515 0.992 Load in System 2.400 1.163 2.667 0.900 Shunt Load in System 0 0 Losses in System 0.095-0.848 Check of Balance 0 0 Table 18: Night time off-peak case with revised RE generation summary Generator Solution Integrating PV Solar and Wind generation with the TEC electric system 56

Name Type Rated MVA MW Mvar MVA PF Vpu Deg GS-1 Sw 2.000 0.246-0.090 0.262-0.939 1.000 0.00 GS-7 PQG 2.700 0.250 0.125 0.280 0.894 1.002-0.47 INV_WND_RAE PQG 4.500 2.000 0.280 2.020 0.990 1.053 0.3 In this revised night time case, a total of 2000 kw of wind generation was modeled taking the conventional generation output down to 496 kw. With this level of wind penetration the system experienced no feeder or transformer overloads as shown in Table 19. Table 19: Night time off-peak case with revised RE feeder and transformer loading summary From Bus Feeder To Bus Branch/Xfmr Name Rated Amps Load Amperes % A-107 A-108 A-2 295 9.1 3.10% A-107 A-PC1 A-3 195 7.9 4.10% A-108 A-108_A A-4 195 1.8 0.90% A-108 A-129_A A-5 195 0.8 0.40% A-108 A-108_B A-7 295 4.9 1.70% A-108_B A-109 A-8 295 2.6 0.90% A-109 A-303 A-9 295 1.8 0.60% A-129 A-129_A A-6 295 0.9 0.30% A-PC1 A-PC3 A-10 195 4.5 2.30% A-PC1 A-PC2 A-13 145 0.8 0.60% A-PC3 A-PC3_B A-11 145 1.8 1.20% A-PC3_B A-PC3_C A-12 195 0.9 0.50% Integrating PV Solar and Wind generation with the TEC electric system 57

From Bus To Bus Branch/Xfmr Name Rated Amps Load Amperes % AC-101 AC-103 AC-4 295 19.1 6.50% AC-102 AC-102_A AC-3 295 0 0.00% AC-103 AC-126 AC-6 295 11.9 4.00% AC-103 AC-102 AC-5 230 2.7 1.20% AC-112 AC-112_B AC-13 295 5.8 2.00% AC-112_A AC-112 AC-12 295 6.7 2.30% AC-112_B AC-130_A AC-14 230 5.8 2.50% AC-118 AC-132_A AC-18 230 1.7 0.70% AC-126 AC-135_A AC-7 230 0.8 0.40% AC-126 AC-112_A AC-11 230 6.6 2.90% AC-130 AC-130_B AC-16 295 4.4 1.50% AC-130_A AC-130 AC-15 295 5.8 2.00% AC-130_B AC-118 AC-17 230 4.4 1.90% AC-132 AC-132_B AC-16B 230 0.1 0.00% AC-132_A AC-132 AC-19 230 1.8 0.80% AC-135 AC-135_B AC-9 295 0 0.00% AC-135_A AC-135 AC-8 295 0.9 0.30% BUS-1 AC-101 AC-1 295 20.8 7.10% BUS-1 EC-119 EC-1 295 20 6.80% BUS-2 WC-110 WC-1 295 80.2 27.20% BUS-2 BUS-1 PS BUS TIE 541 29 5.40% BUS-2 A-107 A-1 195 17.5 9.00% BUS-2 S-105 S-1 295 17.9 6.10% BUS-2 CL-128 CL-1 295 30.7 10.40% CL-128 CL-128_A CL-2 195 0.9 0.50% CL-128 CL-400_A CL-3 295 29.9 10.10% Integrating PV Solar and Wind generation with the TEC electric system 58

From Bus To Bus Branch/Xfmr Name Rated Amps Load Amperes % CL-205 CL-215 CL-39 189 2.3 1.20% CL-205 CL-213 CL-40 295 3.5 1.20% CL-206 CL-220_A CL-32 195 11.9 6.10% CL-207 CL-208_A CL-22 195 15.7 8.10% CL-208 CL-208_B CL-24 295 14.9 5.00% CL-208_A CL-208 CL-23 295 15.8 5.40% CL-208_B CL-219_A CL-25 195 14.9 7.60% CL-209 CL-214_A CL-18 195 18 9.20% CL-210 CL-221_A CL-14 195 21.5 11.00% CL-210 CL-313 CL-7 195 6.4 3.30% CL-211 CL-211_B CL-30 295 13.2 4.50% CL-211_A CL-211 CL-29 295 14.1 4.80% CL-211_B CL-206 CL-31 195 13.2 6.80% CL-213 CL-212 CL-41 295 1.7 0.60% CL-214 CL-214_B CL-20 295 17.3 5.90% CL-214_A CL-214 CL-19 295 18.2 6.20% CL-214_B CL-207 CL-21 295 17.3 5.90% CL-218 CL-218_B CL-37 230 8.4 3.70% CL-218_A CL-218 CL-36 230 9.4 4.10% CL-218_B CL-205 CL-38 195 8.5 4.30% CL-219 CL-219_B CL-27 295 14.1 4.80% CL-219_A CL-219 CL-26 295 15 5.10% CL-219_B CL-211_A CL-28 195 14.1 7.20% CL-220 CL-220_B CL-34 295 9.2 3.10% CL-220_A CL-220 CL-33 295 12 4.10% CL-220_B CL-218_A CL-35 195 9.2 4.70% Integrating PV Solar and Wind generation with the TEC electric system 59

From Bus To Bus Branch/Xfmr Name Rated Amps Load Amperes % CL-221_A CL-221 CL-15 295 21.5 7.30% CL-221_B CL-221 CL-16 295 19.8 6.70% CL-221_B CL-209 CL-17 195 19.8 10.10% CL-312 CL-312_C CL-12 295 0 0.00% CL-312 CL-312_A CL-9 295 5.9 2.00% CL-312_B CL-311 CL-11 195 1.3 0.70% CL-312_B CL-312 CL-10 295 1.3 0.50% CL-313 CL-312_A CL-8 195 5.7 2.90% CL-400_A CL-400 CL-4 195 30.5 15.60% CL-400_B CL-210 CL-6 295 27.9 9.50% CL-400_B CL-400 CL-5 195 27.9 14.30% EC-119 EC-120 EC-2 295 19.2 6.50% EC-120 EC-131 EC-3 195 2.7 1.40% EC-120 EC-121 EC-4 295 15 5.10% EC-121 EC-133_A EC-5 195 12.6 6.40% EC-122 EC-123_A EC-9 195 10.1 5.20% EC-123 EC-123_B EC-114 295 7.7 2.60% EC-123_A EC-123 EC-10 295 10.1 3.40% EC-123_B EC-124 EC-12 195 7.7 3.90% EC-124 EC-134_A EC-13 195 6.9 3.50% EC-133 EC-133_B EC-7 295 11.7 4.00% EC-133_A EC-133 EC-6 295 12.6 4.30% EC-133_B EC-122 EC-8 195 11.7 6.00% EC-134 EC-134_B EC-15 295 6 2.00% EC-134_A EC-134 EC-14 295 6.8 2.30% EC-134_B EC-202 EC-16 195 6 3.10% Integrating PV Solar and Wind generation with the TEC electric system 60

From Bus To Bus Branch/Xfmr Name Rated Amps Load Amperes % EC-202 EC-216_A EC-17 195 4.3 2.20% EC-203 EC-204 EC-24 295 1.3 0.50% EC-216 EC-216_B EC-19 295 3.5 1.20% EC-216_A EC-216 EC-18 295 4.3 1.50% EC-216_B EC-217_A EC-20 195 3.5 1.80% EC-217 EC-217_B EC-22 295 2.7 0.90% EC-217_A EC-217 EC-21 295 3.5 1.20% EC-217_B EC-203 EC-23 195 2.7 1.40% S-105 S-125 S-2 195 15.9 8.20% S-111 S-111_B S-17 295 0 0.00% S-111_A S-111 S-16 295 1.8 0.60% S-113 S-116_A S-12 230 4.4 1.90% S-114 S-117_A S-8 230 6.8 3.00% S-115 S-115_B S-6 230 8.5 3.70% S-115_A S-115 S-5 230 11.2 4.90% S-115_B S-114 S-7 230 8.5 3.70% S-116 S-116_B S-14 295 1.7 0.60% S-116_A S-116 S-13 295 4.4 1.50% S-116_B S-111_A S-15 230 1.7 0.70% S-117 S-117_B S-10 295 6.1 2.10% S-117_A S-117 S-9 295 7 2.40% S-117_B S-113 S-11 230 6.1 2.60% S-125 S-115_A S-4 230 11.1 4.80% S-125 S-127 S-3 195 2.7 1.40% WC-110 WC-302_A WC-3 230 81 35.20% WC-110 WC-301 WC-2 195 0.3 0.10% Integrating PV Solar and Wind generation with the TEC electric system 61

From Bus To Bus Branch/Xfmr Name Rated Amps Load Amperes % WC-302 WC-315 WC-5 295 82.6 28.00% WC-302 WC-302_A WC-4 295 80.9 27.40% WC-304 WC-318 WC-7 295 3.2 1.10% WC-304 WC-305_A WC-9 195 86.6 44.40% WC-305 WC-305_B WC-11 295 89.1 30.20% WC-305_A WC-305 WC-10 295 86.6 29.30% WC-305_B WC-306 WC-12 195 89.1 45.70% WC-306 WC-307 WC-13 195 91.2 46.80% WC-307 WC-308 WC-15 195 93.5 48.00% WC-307 WC-316 WC-14 295 0.8 0.30% WC-308 WC-309 WC-16 195 95.6 49.00% WC-309 WC-321_A WC-17 295 99.2 33.60% WC-309 WC-314_A WC-21 195 2.1 1.10% WC-314 WC-314_B WC-23 230 1.3 0.60% WC-314_A WC-314 WC-22 230 2.1 0.90% WC-314_B WC-310 WC-24 195 1.3 0.70% WC-315 WC-304 WC-6 295 83.4 28.30% WC-318 WC-319 WC-8 295 2.6 0.90% WC-321 WC-321_B WC-19 295 1 0.30% WC-321_A WC-321 WC-18 295 99.2 33.60% WC-321_B WC-317 WC-20 295 0.9 0.30% Transformer TR-1 BUS-1 GS-1 2500 262 10.50% TR-2 GS-2 BUS-1 2500 0 0.00% TR-3 GS-4 BUS-1 750 0 0.00% TR-4 GS-5 BUS-1 750 0 0.00% Integrating PV Solar and Wind generation with the TEC electric system 62

Bus 1 Frequency (Hz) DNV KEMA Energy & Sustainability From Bus To Bus Branch/Xfmr Name Rated Amps Load Amperes % TR-7 NZ-GS BUS-1 2500 0 0.00% TX-RE-RAEMARU WC-321 WND_1_RAE 4500 2018 44.80% 4.2.1 Night case with revised RE Power Quality Discussion In this revised night time case, the originally proposed wind locations were removed and a total of 2000 kw of wind generation was modeled at Raemaru; taking the conventional generation output down to 496 kw. Grid performance appears acceptable with both the initial and final Night-time RE scenarios with no discernible power quality issues. 4.2.2 Night case with revised RE Dynamic Stability Analysis Disturbance: Loss of largest diesel unit (GS-7); dispatched at 250 kw in the day time peak case. Figure 35: Off-Peak with revised RE; Loss of GS-7 (No fault); Frequency Response Off-Peak Load with Wind Loss of largest unit (250 kw) 50.05 50 49.95 49.9 49.85 49.8 49.75 49.7 49.65 0 1 2 3 4 5 6 7 8 9 10 Integrating PV Solar and Wind generation with the TEC electric system 63

Figure 36: Off-Peak with revised RE; Loss of GS-7 (No fault); Voltage Response Bus 1 Voltage (pu) 1.005 1 0.995 0.99 0.985 0.98 0.975 0.97 0.965 Off-Peak Load with Wind Loss of largest unit (250 kw) 0 1 2 3 4 5 6 7 8 9 10 Similar to the revised day time peak case in Section 4.1.1, due to the reduced output of the diesel units, loss of one diesel does not have a large dynamic impact on the island power system. The frequency dips to 49.8 Hz and recovers as shown in Figure 35, while the voltage momentarily drops to 0.98 pu but quickly returns to 1.0 pu; as can be seen in Figure 36. Integrating PV Solar and Wind generation with the TEC electric system 64

Bus 1 Voltage (pu) DNV KEMA Energy & Sustainability Disturbance: Apply a 3-phase fault to the West Coast 11 kv Feeder at first bus outside of Avatiu Valley Power Station then clear the fault by opening the feeder breaker. Figure 37: West Coast Feeder fault; Frequency Response Off-Peak Load with Wind West Coast Feeder fault Bus 1 Frequency (Hz) 50.2 50 49.8 49.6 49.4 49.2 49 0 1 2 3 4 5 6 7 8 9 Figure 38: West Coast Feeder fault; Voltage Response 1.2 Off-Peak Load with Wind West Coast Feeder fault 1 0.8 0.6 0.4 0.2 0 0 1 2 3 4 5 6 7 8 9 Figure 37 above shows the frequency response for a close-in fault on the West Coast Feeder and subsequent clearing of the fault. Raemaru wind generation is interconnected to this feeder so this sensitivity results in tripping of this generation along with the feeder and presents the most extreme case Integrating PV Solar and Wind generation with the TEC electric system 65

for dynamic analysis. The frequency drops below 49.4 Hz but returns to 50 Hz and remains stable. Figure 38 shows the voltage response for the same contingency condition. Voltage at the Avatiu Valley Power Station bus is depressed to near zero during the fault, but recovers quickly and returns to 1.0 pu. Integrating PV Solar and Wind generation with the TEC electric system 66

Bus Voltage (pu) DNV KEMA Energy & Sustainability Disturbance: Loss of Raemaru Wind Generation; dispatched at 2000 kw in the night time off-peak model. Figure 39: Loss of Raemaru 2000 kw Wind (No fault); Frequency Response Off-Peak Load with Wind Loss of Raemaru Wind Generation Bus Frequency (Hz) 50.2 50 49.8 49.6 49.4 49.2 49 48.8 48.6 0 1 2 3 4 5 6 7 8 9 10 Figure 40: Loss of Raemaru 2000 kw Wind (No fault); Voltage Response 1.06 1.04 1.02 1 0.98 0.96 0.94 0.92 Off-Peak Load with Wind Loss of Raemaru Wind Generation 0 1 2 3 4 5 6 7 8 9 10 Integrating PV Solar and Wind generation with the TEC electric system 67

Figure 39 above shows the frequency response at the customer transformer closest to the interconnection with the Raemaru wind generation. The frequency at this customer bus dips to 49.1 Hz and recovers within an acceptable timeframe. Figure 40 shows the voltage response for the same contingency condition. Voltage at the PV interconnection bus drops below 0.97 pu but recovers and remains stable at greater than 0.98 pu. Figure 41: Loss of Raemaru 2000 kw Wind (No fault); Diesel Generator Real Power Output Off-Peak Load with Wind Loss of Raemaru Wind Generation Generation Real Power Output 1600 1400 1200 1000 800 600 400 200 0 0 1 2 3 4 5 6 7 8 9 10 GS-1 [kw] GS-7 [kw] Integrating PV Solar and Wind generation with the TEC electric system 68

Generation Speed (RPM) DNV KEMA Energy & Sustainability Figure 42: Loss of Raemaru 2000 kw Wind (No fault); Diesel Generator Reactive Power Output Off-Peak Load with Wind Loss of Raemaru Wind Generation 300 Generation Reactive Power Output 250 200 150 100 50 0 0 1 2 3 4 5 6 7 8 9 10 GS-1 [kvar] GS-7 [kvar] Figure 43: Loss of Raemaru 2000 kw Wind (No fault); Diesel Generator Rotations Per Minute 3010 3000 2990 2980 2970 2960 2950 2940 2930 2920 Off-Peak Load with Wind Loss of Raemaru Wind Generation 0 1 2 3 4 5 6 7 8 9 10 GS-1 [RPM] GS-7 [RPM] The three figures above show the real and reactive power as well as the speed of both diesel generators during the loss of the large wind installation at Raemaru. No significant issues are detected. Integrating PV Solar and Wind generation with the TEC electric system 69

4.3 Revised Day case with maximum PV generation In this scenario, the day time base case was revised to include the maximum amount of PV generation. Wind resources were removed from the model and PV sites were modeled based on the priority ranking shown in Table 20 below. Wind generation in the previously revised day time peak model (Section 4.1) total 700 kw. In this scenario that generation will be replaced with an additional 700 kw of PV resources by including the West of Parking Area PV site to the model. The Panama Dump site will remain dispatched at 1000 kw. Table 20: Potential Large-scale PV Installations 16 The following are the assumptions used in the power flow model for the Project Scenario described in the following section: Model from first round steady-state and dynamic stability analysis Day-time Peak Case with Wind and PV. Load is 4,830 kw on-peak (DAY) load. 16 Source: Potential Sites for Grid Assessment J000898 Rarotonga Large Scale PV; Southern Perspectives; 12/12/2012 Integrating PV Solar and Wind generation with the TEC electric system 70

Two diesel generators were assumed to be running; 2 units online dispatched at 234 kw and 250 kw, respectively. At least one diesel unit is needed to provide a regulated voltage and frequency source for the island. Running two diesel units allows either diesel unit to continue regulating voltage and frequency if the other unit trips off line. This is typically referred to as an n-1 resource criterion. Three wind generation sites at the Hospital, Matavera and West Wigmores sites were removed from the model. Seven PV sites were modeled for a total of 4478 kw as shown in Table 21 Table 21: Maximum PV Renewable Resources Summary Site DRG Type Capacity (kw) Assumed Output (kw) FDR Sun City PV 2000 2000 SEAPORT TSA Stadium PV 195 195 AIRPORT Telecom PV 210 210 CROSS LINE Cultural Center PV 155 155 EAST COAST Airport Hanger PV 120 120 AIRPORT Airport Terminal PV 98 98 AIRPORT Panama Dump PV 1,300 1,000 SEAPORT West Parking PV 700 700 AIRPORT Table 22 below shows the generation and load summary for the revised model and Table 15 shows the generation dispatch conditions modeled for the PV project and diesel generators. Integrating PV Solar and Wind generation with the TEC electric system 71

Table 22: Day time peak case with maximum PV system summary Total MW Mvar MVA pf Generation in System 4.968 1.565 5.209 0.954 Load in System 4.830 2.339 5.367 0.900 Shunt Load in System 0 0 Losses in System 0.165-0.778 Check of Balance 0 0 Table 23: Day time peak case with maximum PV generation summary Name Generator Type Solution Rated MVA MW Mvar MVA PF Vpu Deg GS-1 Sw 2.000 0.242 0.741 0.779 0.310 1.020 0 GS-7 PQG 2.700 0.250 0.150 0.292 0.857 0.996-0.24 INV_CLTR_PV PQG 0.200 0.155 0.022 0.157 0.990 0.996.06 INV_DUMP_PV1 PQG 0.750 0.500 0.142 0.520 0.962 1.029 0.37 INV_DUMP_PV2 PQG 0.750 0.500 0.142 0.520 0.962 1.029 0.37 INV_SUNCTY_PV1 PQG 1.500 1.000 0.142 1.011 0.989 1.033 0.50 INV_SUNCTY_PV2 PQG 1.500 1.000 0.142 1.011 0.989 1.033 0.50 INV_TLCM_PV PQG 0.250 0.210 0.032 0.212 0.989 1.001 0.23 INV_TRMHNGPRK_PV PQG 1.250 0.918 0.033 0.928 0.990 1.008 0.31 INV_TSA_PV PQG 0.250 0.195 0.028 0.197 0.990 1.008 0.31 In this revised day time case, a total of 4,478 kw of PV generation was modeled taking the conventional generation output down to 490 kw. With this maximum level of PV penetration the system experienced no feeder or transformer overloads as shown in Table 24. Integrating PV Solar and Wind generation with the TEC electric system 72

Table 24: Day time peak case with maximum PV feeder and transformer loading summary From Bus To Bus Branch/Xfmr Name Feeder Rated Amps Load Amperes % A-107 A-PC1 A-3 195 33 16.90% A-107 A-108 A-2 295 8.8 3.00% A-108 A-129_A A-5 195 1.7 0.90% A-108 A-108_A A-4 195 6.9 3.50% A-108 A-108_B A-7 295 10.1 3.40% A-108_B A-109 A-8 295 5 1.70% A-109 A-303 A-9 295 3.3 1.10% A-129 A-129_A A-6 295 1.8 0.60% A-PC1 A-PC2 A-13 145 1.7 1.20% A-PC1 A-PC3 A-10 195 9.1 4.70% A-PC3 A-PC3_B A-11 145 3.6 2.50% A-PC3_B A-PC3_C A-12 195 1.8 0.90% AC-101 AC-103 AC-4 295 28.6 9.70% AC-102 AC-102_A AC-3 295 0 0.00% AC-103 AC-126 AC-6 295 24.5 8.30% AC-103 AC-102 AC-5 230 6.1 2.60% AC-112 AC-112_B AC-13 295 11.9 4.00% AC-112_A AC-112 AC-12 295 13.8 4.70% AC-112_B AC-130_A AC-14 230 12 5.20% AC-118 AC-132_A AC-18 230 3.6 1.60% AC-126 AC-112_A AC-11 230 13.7 5.90% AC-126 AC-135_A AC-7 230 1.7 0.70% AC-130 AC-130_B AC-16 295 9 3.10% Integrating PV Solar and Wind generation with the TEC electric system 73

From Bus To Bus Branch/Xfmr Name Rated Amps Load Amperes % AC-130_A AC-130 AC-15 295 12 4.10% AC-130_B AC-118 AC-17 230 9.1 3.90% AC-132 AC-132_B AC-16B 230 0.1 0.00% AC-132_A AC-132 AC-19 230 3.6 1.60% AC-135 AC-135_B AC-9 295 0 0.00% AC-135_A AC-135 AC-8 295 1.8 0.60% BUS-1 EC-119 EC-1 295 33.4 11.30% BUS-1 AC-101 AC-1 295 32.5 11.00% BUS-2 CL-128 CL-1 295 65.3 22.10% BUS-2 BUS-1 PS BUS TIE 541 49.6 9.20% BUS-2 A-107 A-1 195 23.6 12.10% BUS-2 WC-110 WC-1 295 45.3 15.30% BUS-2 S-105 S-1 295 119.9 40.70% CL-128 CL-400_A CL-3 295 63.6 21.60% CL-128 CL-128_A CL-2 195 1.8 0.90% CL-205 CL-215 CL-39 189 4.8 2.50% CL-205 CL-213 CL-40 295 7.4 2.50% CL-206 CL-220_A CL-32 195 25.1 12.90% CL-207 CL-208_A CL-22 195 33.4 17.10% CL-208 CL-208_B CL-24 295 31.6 10.70% CL-208_A CL-208 CL-23 295 33.5 11.40% CL-208_B CL-219_A CL-25 195 31.6 16.20% CL-209 CL-214_A CL-18 195 38.5 19.70% CL-210 CL-313 CL-7 195 13.6 7.00% CL-210 CL-221_A CL-14 195 45.8 23.50% CL-211 CL-211_B CL-30 295 28 9.50% Integrating PV Solar and Wind generation with the TEC electric system 74

From Bus To Bus Branch/Xfmr Name Rated Amps Load Amperes % CL-211_A CL-211 CL-29 295 29.9 10.10% CL-211_B CL-206 CL-31 195 28 14.40% CL-213 CL-212 CL-41 295 3.7 1.30% CL-214 CL-214_B CL-20 295 36.9 12.50% CL-214_A CL-214 CL-19 295 38.7 13.10% CL-214_B CL-207 CL-21 295 36.9 12.50% CL-218 CL-218_B CL-37 230 17.8 7.70% CL-218_A CL-218 CL-36 230 19.7 8.60% CL-218_B CL-205 CL-38 195 17.8 9.10% CL-219 CL-219_B CL-27 295 29.9 10.10% CL-219_A CL-219 CL-26 295 31.8 10.80% CL-219_B CL-211_A CL-28 195 29.9 15.30% CL-220 CL-220_B CL-34 295 19.6 6.60% CL-220_A CL-220 CL-33 295 25.3 8.60% CL-220_B CL-218_A CL-35 195 19.6 10.00% CL-221_A CL-221 CL-15 295 45.9 15.60% CL-221_B CL-221 CL-16 295 42.1 14.30% CL-221_B CL-209 CL-17 195 42.1 21.60% CL-312 CL-312_C CL-12 295 0 0.00% CL-312 CL-312_A CL-9 295 12.2 4.10% CL-312_B CL-311 CL-11 195 2.8 1.50% CL-312_B CL-312 CL-10 295 2.8 1.00% CL-313 CL-312_A CL-8 195 12 6.10% CL-400_A CL-400 CL-4 195 64.9 33.30% CL-400_B CL-210 CL-6 295 59.4 20.10% CL-400_B CL-400 CL-5 195 59.4 30.50% Integrating PV Solar and Wind generation with the TEC electric system 75

From Bus To Bus Branch/Xfmr Name Rated Amps Load Amperes % EC-119 EC-120 EC-2 295 32.3 11.00% EC-120 EC-121 EC-4 295 31.5 10.70% EC-120 EC-131 EC-3 195 3.4 1.80% EC-121 EC-133_A EC-5 195 26.3 13.50% EC-122 EC-123_A EC-9 195 21.1 10.80% EC-123 EC-123_B EC-114 295 15.7 5.30% EC-123_A EC-123 EC-10 295 21.2 7.20% EC-123_B EC-124 EC-12 195 15.7 8.10% EC-124 EC-134_A EC-13 195 14 7.20% EC-133 EC-133_B EC-7 295 24.6 8.30% EC-133_A EC-133 EC-6 295 26.3 8.90% EC-133_B EC-122 EC-8 195 24.7 12.60% EC-134 EC-134_B EC-15 295 12.3 4.20% EC-134_A EC-134 EC-14 295 14 4.70% EC-134_B EC-202 EC-16 195 12.4 6.40% EC-202 EC-216_A EC-17 195 8.9 4.60% EC-203 EC-204 EC-24 295 2.8 1.00% EC-216 EC-216_B EC-19 295 7.2 2.40% EC-216_A EC-216 EC-18 295 8.9 3.00% EC-216_B EC-217_A EC-20 195 7.2 3.70% EC-217 EC-217_B EC-22 295 5.5 1.90% EC-217_A EC-217 EC-21 295 7.2 2.40% EC-217_B EC-203 EC-23 195 5.6 2.90% S-105 S-125 S-2 195 124.1 63.60% S-111 S-111_B S-17 295 0 0.00% S-111_A S-111 S-16 295 3.6 1.20% Integrating PV Solar and Wind generation with the TEC electric system 76

From Bus To Bus Branch/Xfmr Name Rated Amps Load Amperes % S-113 S-116_A S-12 230 8.7 3.80% S-114 S-117_A S-8 230 89.4 38.90% S-115 S-115_B S-6 230 138.7 60.30% S-115_A S-115 S-5 230 133.5 58.00% S-115_B S-114 S-7 230 138.6 60.30% S-116 S-116_B S-14 295 3.4 1.20% S-116_A S-116 S-13 295 8.7 3.00% S-116_B S-111_A S-15 230 3.4 1.50% S-117 S-117_B S-10 295 12.1 4.10% S-117_A S-117 S-9 295 89.4 30.30% S-117_B S-113 S-11 230 12.1 5.30% S-125 S-115_A S-4 230 133.5 58.10% S-125 S-127 S-3 195 5.4 2.80% WC-110 WC-301 WC-2 195 0.2 0.10% WC-110 WC-302_A WC-3 230 43.3 18.80% WC-302 WC-315 WC-5 295 39.8 13.50% WC-302 WC-302_A WC-4 295 43.5 14.80% WC-304 WC-305_A WC-9 195 31.2 16.00% WC-304 WC-318 WC-7 295 7 2.40% WC-305 WC-305_B WC-11 295 25.7 8.70% WC-305_A WC-305 WC-10 295 31.2 10.60% WC-305_B WC-306 WC-12 195 25.8 13.20% WC-306 WC-307 WC-13 195 21.2 10.90% WC-307 WC-308 WC-15 195 15.9 8.20% WC-307 WC-316 WC-14 295 1.7 0.60% WC-308 WC-309 WC-16 195 11.4 5.90% Integrating PV Solar and Wind generation with the TEC electric system 77

From Bus To Bus Branch/Xfmr Name Rated Amps Load Amperes % WC-309 WC-314_A WC-21 195 4.6 2.40% WC-309 WC-321_A WC-17 295 3.4 1.10% WC-314 WC-314_B WC-23 230 2.9 1.20% WC-314_A WC-314 WC-22 230 4.7 2.10% WC-314_B WC-310 WC-24 195 2.9 1.50% WC-315 WC-304 WC-6 295 38.1 12.90% WC-318 WC-319 WC-8 295 5.5 1.90% WC-321 WC-321_B WC-19 295 1.7 0.60% WC-321_A WC-321 WC-18 295 3.4 1.20% WC-321_B WC-317 WC-20 295 1.7 0.60% Transformer TR-1 BUS-1 GS-1 2500 648 25.90% TR-2 GS-2 BUS-1 2500 0 0.00% TR-3 GS-4 BUS-1 750 0 0.00% TR-4 GS-5 BUS-1 750 0 0.00% TR-7 NZ-GS BUS-1 2500 0 0.00% TX-RE-CLTR_PV EC-131 CLTR_PV_ACBUS 250 156 62.50% TX-RE-DUMP_PV S-114 DUMP_PV_ACBUS 1500 1037 69.10% TX-RE-SUNCTY_PV S-117 SUNCTY_PV_ACBUS 2500 2018 80.70% TX-RE-TLCM_PV AC-102 TLCM_PV_ACBUS 250 212 84.70% TX-RE-TRMHNGP_PV A-PC1 TRMHNGP_PV_ACBUS 1250 925 74.00% TX-RE-TSA_PV A-108_A TSA_PV_ACBUS 250 196 78.50% Integrating PV Solar and Wind generation with the TEC electric system 78

4.3.1 Day case with maximum PV Power Quality Discussion Additional PV projects were added to maximize the PV output, and wind generation was removed from this final Day-time case as summarized above. Grid performance appears acceptable for the maximum PV scenario, except for flicker concerns. Most of the currently proposed PV project sites in Table 22 are located near the northwest coast of the island, and more or less along a line from the airport area to the cultural center area. Given this narrow geographical distribution these sites could all be affected concurrently by cloud fronts passing over that part of the island. The impact of passing cloud cover can cause PV to ramp down from peak output to 30% of rated output, or lower. This causes voltage deviations (flicker) on the system, which can present power quality problems for customers. As discussed in the following section, KEMA s grid analysis simulated such conditions and determined that the voltage drop would exceed the IEEE guidelines shown in Figure 44. Figure 44: Voltage Flicker curve from IEEE 141 & 519 17 4.3.2 Day case with maximum PV Dynamic Stability Analysis Disturbance: Loss of largest diesel unit (GS-7); dispatched at 250 kw in the day time peak case. 17 Based on IEC Standard 1000-4-15 Integrating PV Solar and Wind generation with the TEC electric system 79