2018 A.T. Procopiou - The University of Melbourne MIE Symposium, December 2018 1 Increasing PV Hosting Capacity in Distribution Networks: Challenges and Opportunities Dr Andreas T. Procopiou Research Fellow in Smart Grids andreas.procopiou@unimelb.edu.au www.andreasprocopiou.com The University of Melbourne Melbourne Institute of Energy Symposium 12 th December 2018
Solar PV in Australia Status, installations and cumulative capacity Challenges in PV-rich Distribution Networks Traditional and non-traditional mitigation approaches Understanding Solar PV impacts Smart PV Inverters Embedded Control Functions Increasing PV Hosting Capacity Residential Battery Energy Storage Systems Opportunity for advanced controllers (to manage technical issues) Conclusions Outline 2018 A.T. Procopiou - The University of Melbourne MIE Symposium, December 2018 2
Number of Installations 62% Capacity (kwp) 123% Thousands Solar PV in Australia 20 18 16 14 12 10 8 6 4 2 0 Installations Total Monthly Installations Average Installed Capacity 9 8 7 6 5 4 3 2 1 0 Cumulative Installed Capacity 2018 2.9GW 9.00 8.00 7.00 6.00 5.00 Total Installed Capacity (GW)10.00 2016 0.8GW 2017 1.3GW 2 million Installations 18GW by 2020 Solar PV Status, Australia: Australian PV Institute, 2018. [Online]. Available: https://goo.gl/n8cjff. Accessed on November 2018 2018 A.T. Procopiou - The University of Melbourne MIE Symposium, December 2018 3
2018 A.T. Procopiou - The University of Melbourne MIE Symposium, December 2018 4 Challenges in PV-rich Distribution Networks Bulk Generation Transmission Distribution MV/LV Bulk supply point Voltage Max Min Distance PV Systems Not Generating
2018 A.T. Procopiou - The University of Melbourne MIE Symposium, December 2018 5 Challenges in PV-rich Distribution Networks Bulk Generation Technical issues brought by high penetrations of solar PV significantly reduces hosting capacity of networks Transmission Distribution Congestion Voltage rise MV/LV Bulk supply point Traditional Solutions Network Reinforcement Voltage Max Min Distance PV Systems PV Systems Not Generating Generating
2018 A.T. Procopiou - The University of Melbourne MIE Symposium, December 2018 6 Challenges in PV-rich Distribution Networks Solutions Bulk Generation Transmission Bigger Transformers Larger Cables MV/LV Traditional Solutions Network Reinforcement Bulk supply point Voltage Max Min Distance
2018 A.T. Procopiou - The University of Melbourne MIE Symposium, December 2018 7 Challenges in PV-rich Distribution Networks Solutions Leveraging existing assets to manage technical issues and increase hosting capacity Bulk Generation Reinforcement Alternative Transmission Distribution MV/LV Generation Curtailment Reactive Power Absorption Bulk supply point Non-Traditional Solutions Smart Storage PV inverter and solar capabilities PV (reduce household exports) Max Min Distance
Understanding Solar PV Impacts Completed project Solar PV Penetration and HV-LV Network Impacts Project Real Victorian 22kV HV feeder Strong semi-urban 30km of conductors 79 distribution transformers Realistically Modelled LV Networks Australian Design Principles 175 LV feeders 4612 residential customers Stochastic Analyses - Monte Carlo Summer (December February) Varying locations and sizes (using regional PV stats) Smart meter demand and PV generation PV penetration increments of 10% (0-100%) % of customers with PV systems 1 1 A. Navarro, L.F. Ochoa, Probabilistic impact assessment of low carbon technologies in LV distribution systems, IEEE Trans. on Power Systems, May 2016 (10.1109/TPWRS.2015.2448663) 2018 A.T. Procopiou - The University of Melbourne MIE Symposium, December 2018 8
2018 A.T. Procopiou - The University of Melbourne MIE Symposium, December 2018 9 Understanding Solar PV Impacts Stochastic Impact Analyses Completed project Solar PV Penetration and HV-LV Network Impacts Project LV Voltage Issues HV Conductors Congestion Default Volt-Watt settings (AS/NSZ 4777.2:2015) not adequate to manage issues Hosting Capacity: 20% PV Penetration * Off-load taps at nominal position (3) and Volt-Watt function as per AS/NSZ 4777.2:2015
2018 A.T. Procopiou - The University of Melbourne MIE Symposium, December 2018 10 Source: www.abb.com Smart PV Inverters Embedded Controllability Source: www.solaredge.com Source: www.sma-australia.com.au Embedded with power control functions Volt-Watt Volt-var Fixed PF Watt-PF Power Limit Embedded with communication interfaces
2018 A.T. Procopiou - The University of Melbourne MIE Symposium, December 2018 11 Smart PV Inverters Embedded Control Functions 1. Active Power Limit Function 2. Volt-Watt Control Function Inverter Power Priority Watt Priority Limited Q 3. Volt-var Control Function 4. Watt-PF Control Function (a) Var Watt Priority
Max Watt Output (% of max output) % of available Vars Smart PV Inverters Control Function Examples Example settings used for demonstration purposes: Volt-Watt 100% 75% Volt-var 100% 50% 50% 0% 0.9 0.94 0.98 1.02 1.06 1.1 25% -50% 0% 1.05 1.06 1.07 1.08 1.09 1.1 1.11 1.12 Voltage (p.u.) -100% Voltage (p.u.) A. Procopiou, Active Management of PV-Rich Low Voltage Networks, PhD Thesis, The Univ. of Manchester, 2017 (https://www.escholar.manchester.ac.uk/item/?pid=uk-ac-man-scw:310939) 2018 A.T. Procopiou - The University of Melbourne MIE Symposium, December 2018 12
2018 A.T. Procopiou - The University of Melbourne MIE Symposium, December 2018 13 Smart PV Inverters Voltage Issues 60% PV Penetration on the Australian HV-LV Network BAU Volt-Watt Volt-var 34% 0% 34% 12% curtailment Volt-Watt control effective at the expense of energy curtailment Volt-var control ineffective due to limited Q when needed
2018 A.T. Procopiou - The University of Melbourne MIE Symposium, December 2018 14 Smart PV Inverters Voltage Issues 60% PV Penetration on the Australian HV-LV Network BAU Volt-Watt Volt-var Volt-var (oversized or Var priority) 34% 0% 34% 0% 12% curtailment Volt-Watt control effective at the expense of energy curtailment Volt-var control ineffective due to limited Q when needed
2018 A.T. Procopiou - The University of Melbourne MIE Symposium, December 2018 15 Smart PV Inverters Thermal Issues 60% Penetration on the Australian HV-LV Network BAU Volt-Watt Volt-var Volt-var (oversized) 5 Txs overloaded 0 Txs overloaded 76 Txs overloaded HIGHER UTILIZATION Curtailment from Volt-Watt eliminates Tx overloads Q from Volt-var creates more overloads
Reactive Power (% Available VARs) Power Factor 2018 A.T. Procopiou - The University of Melbourne MIE Symposium, December 2018 16 Active Power (% Max Power) Smart PV Inverters Control Function Settings and Options 50% 40% 30% 20% 10% 0% 0.85-10% 0.9 0.95 1 1.05 1.1 1.15-20% -30% Volt-var California Hawaii IEEE-Cat A IEEE-Cat B AU/NZ Italy/Austria 120% 100% 80% 60% 40% 20% IEEE/Hawaii AU/NZ Austria Volt-Watt -40% -50% Voltage (p.u.) 0% 1 1.05 1.1 1.15 Voltage (p.u.) 1.02 1.00 0.98 0.96 Watt-PF 0.94 0.92 AU/NZ 0.90 Italy/Austria/Germany 0.88 0% 20% 40% 60% 80% 100% Active Power (% Max Power) (a) Watt Priority (b) Var Priority (c) 10% Oversized with Watt Priority
Reactive Power (% Available VARs) Power Factor 2018 A.T. Procopiou - The University of Melbourne MIE Symposium, December 2018 17 Active Power (% Max Power) 50% 40% 30% 20% 10% -30% -40% -50% California Hawaii IEEE-Cat A IEEE-Cat B AU/NZ Italy/Austria 60% 0% 0.85 0.9 0.95 1 1.05 1.1 1.15-10% Which is more adequate to mitigate 40% issues? -20% 1.02 Volt-var Smart PV Inverters Control Function Settings and Options What settings offer more 0% benefits? 1 1.05 1.1 1.15 Voltage (p.u.) Voltage (p.u.) 120% 100% Extend of additional Hosting Capacity? 80% 20% IEEE/Hawaii AU/NZ Austria Volt-Watt 1.00 0.98 0.96 Significant number of solution options Complex! Watt-PF 0.94 0.92 AU/NZ 0.90 Italy/Austria/Germany 0.88 0% 20% 40% 60% 80% 100% Active Power (% Max Power) (a) Watt Priority (b) Var Priority (c) 10% Oversized with Watt Priority
On going project 2018 A.T. Procopiou - The University of Melbourne MIE Symposium, December 2018 18 Selection of Network Increasing PV Hosting Capacity Hosting Capacity and Impact Solutions Assessment Tool Technical Issues Solution Methods for Increasing PV Hosting Capacity Project Specification of new PV (location, inverter) Real 22kV Feeder (200+ Dist. Tx, 500+ Customers) Hosting Capacity Limit Assessment Location of new PV System Solutions Assessment Primary Substation Distribution Substation Analysis Summary Used by Distribution Network Planners Next Level Hosting Capacity Limitation Analysis Summary
Source: Clean Energy Australia Report 2018 Number of Installations Thousands 2018 A.T. Procopiou - The University of Melbourne MIE Symposium, December 2018 19 Residential Battery Energy Storage Systems Status in Australia Increasing interest of customers in Battery Energy Storage (BES) Store excess of PV generation and use it later Reduce grid imports; hence electricity bills 25 RESIDENTIAL ENERGY STORAGE SYSTEM INSTALLATIONS 20 15 10 three-fold increase from 2016 5 0 2015 2016 2017 Source: Clean Energy Australia Report 2018
Residential Battery Energy Storage Systems Off-the-shelf Operation and the Opportunity Off-the-shelf (OTS) BES operate for the sole benefit of the customer Do not provide benefits to the network 1 OTS Battery Controller BES systems have different control capabilities Opportunity to provide benefits to both network and customers Reduce reverse power flows, hence, network issues Alternative to costly network reinforcements Allow customers reduce electricity bills Household with 5kWp PV system, and 5kW/13.5kWh BES system 1 K. Petrou, L.F. Ochoa, A.T. Procopiou, J. Theunissen, J. Bridge, T. Langstaff, K. Lintern, "Limitations of residential storage in PV-rich distribution networks: An Australian case study" 2018 IEEE Power & Energy Society General Meeting 2018 A.T. Procopiou - The University of Melbourne MIE Symposium, December 2018 20
Residential Battery Energy Storage Systems Residential Storage Controller for the Benefit of Customers and Networks 1 The Developed Battery Controller: Adapts charging power to the PV generation and Demand Reduces reverse power flow during peak generation periods Ensures available capacity by discharging overnight Always supports the demand, throughout the day Adapts to sudden changes in demand and generation No Communication Infrastructure Required Uses Local Measurements and Known Data Local Measurements: PV generation, Demand, SOC Known data: Clear-sky irradiance Reinforcement Alternative 1 A.T. Procopiou, K. Petrou, and L.F. Ochoa, "A controller for photovoltaic generation and energy storage system," Australia Patent 2018904310, 2018. Available: https://goo.gl/vysfmj. 2018 A.T. Procopiou - The University of Melbourne MIE Symposium, December 2018 21
2018 A.T. Procopiou - The University of Melbourne MIE Symposium, December 2018 22 Voltages Residential Battery Energy Storage Systems Network Benefits PV Only Off-the-Shelf (OTS) With Proposed Battery Controller 18% Non-Compliant 10% Non-Compliant No Voltage Issues
2018 A.T. Procopiou - The University of Melbourne MIE Symposium, December 2018 23 Asset Utilization Residential Battery Energy Storage Systems Network Benefits PV Only Off-the-Shelf (OTS) With Proposed Battery Controller With Proposed Battery Controller Lines and TXs Congested Lines and TXs Congested No Congestions
2018 A.T. Procopiou - The University of Melbourne MIE Symposium, December 2018 24 Residential Battery Energy Storage Systems Customer Benefits Customer Grid Dependency - Year Analysis Grid Dependence Index % of demand imported from the grid 100% = Fully dependant to the grid With Proposed Battery Controller 0% = Energy Self-Sufficient
2018 A.T. Procopiou - The University of Melbourne MIE Symposium, December 2018 25 Conclusions 1/2 DNSPs face challenges evaluating the growing penetrations of PV systems Locational and behavioral uncertainties of PV systems Simplified impact analyses are not adequate to cover uncertainties Advanced computational simulation models and techniques are required Detailed time-series analyses (three-phase, MV-LV, daily/seasonal demand/generation) Stochastic assessment (catering for uncertainties) Increasing PV hosting capacity: Leveraging existing assets (cost effective) Smart PV Inverters offer a wide range of solution options Volt-Watt is effective but in expense of curtailment Volt-var might be effective (if capability exists) but exacerbates asset utilization Complexity in identifying the most adequate combination of control and settings Advanced solution assessment tools required
2018 A.T. Procopiou - The University of Melbourne MIE Symposium, December 2018 26 Conclusions 2/2 Increasing PV hosting capacity: Leveraging existing assets (cost effective) Residential BES Systems OTS control strategies (customer benefit oriented) do not increase PV hosting capacity Opportunity for new storage control strategies providing benefits to both: Network (management of technical issues, increasing HC) Customers (reduced grid imports, hence electricity bills) Trade-off between technical performance, customer impacts, practicality, and cost should always be taken into consideration
2018 A.T. Procopiou - The University of Melbourne MIE Symposium, December 2018 27 Thank you! Acknowledgement Mr Kyriacos Petrou PhD Student Prof Luis F. Ochoa Professor in Power Systems
2018 A.T. Procopiou - The University of Melbourne MIE Symposium, December 2018 28 Increasing PV Hosting Capacity in Distribution Networks: Challenges and Opportunities Dr Andreas T. Procopiou Research Fellow in Smart Grids andreas.procopiou@unimelb.edu.au www.andreasprocopiou.com The University of Melbourne Melbourne Institute of Energy Symposium 12 th December 2018