Grid Integration of Distributed

1 Grid Integration of Distributed Generation and Storage Tom Key, Technical Executive EPRI tkey@epri.com Tuesday, July 29, 2014 National Harbor, MD

The Electric Power System

Variable solar and wind generation A US Future of 302 GW PV by 2030? Generation in Germany Week - April 14-20, 2014 60 GW DOE SunShot Vision Study, Released February 2012 Is the grid ready? Mo Tu We Th Fr Sa Su

Steps for Integrating DG? Understanding PV Plant Performance Determine Feeder Hosting Capabilities Develop Fast Track Screening Deploy Inverter Grid Support Current Research Activities

EPRI SolarTAC Testing EPRI SolarTAC 1.2 Current (A) 1.0 1 0.8 0.6 0.4 02 0.2 0.0 0 0.5 20 40 60 80 Voltage (V) 100 Seasonal Performance Factors for SolarTAC 1.1 Suntech Trina 60 Weighted Average Module Temperature ( C) ( C) for Suntech Helios (Dec 09, 2013)... Poly1 Poly2 Mono 12:00 Local Time CdTe 1 15:00 CIGS1 CIGS2 HIT 50 First Solar Solar Frontier 0.9 40 Stion Panasonic 30 0.8 20 0.7 0.6 10 Q1(Jan-Mar) 2013 Q2(Apr-Jun) 2013 Q3(Jul-Sep) 2013 Season Q4(Oct-Dec) 2013 0 Tempe erature ( C) 0 Perform mance Factor No rm aliz ed O utput Normalized I V Curves 1.4

Seasonal PV Daily Output Profile Shows max, min, median, and middle 50% range for a Georgia Site 0.8 0.6 0.4 0.2 1 Winter (Jan-Mar) 0 10 15 20 Normalized Power 1 Summer (Jul-Sep) 0.8 0.6 068 0.68 1 Spring (Apr-Jun) 018 0.18 span 0.8 span 0.3 span 0.6 0.4 0.2 0 0.8 0.6 1 Fall (Oct-Dec) 10 15 20 0.58 span 04 0.4 0.2 0 10 15 20 Hour of Day 04 0.4 0.2 0 10 15 20

Solar Resource Calendar 1MW AC Output Power December 2011: Tennessee 1MW PV System Power Sun Mon Tue Wed Thu Fri Sat 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 Calendar profiles are 1 minute averages derived from 1 sec data

Categories for Daily Variability Conditions Sandia s variability index (VI) and clearness index (CI) to classify days VI > 10 Clear Sky POA Irradiance Measured POA Irradiance High VI < 2 CI 0.5 Clear Moderate 5 VI < 10 VI < 2 Overcast Mild 2 VI < 5 CI 0.5

Geographical and Seasonal PV Variability Variability Conditions: NM Variability Conditions: NJ Percentage of Days (%) 100 80 60 40 20 0 Q2 2012 Q3 2012 Q4 2012 Q1 2013 Season f Days (%) Percentage of 100 80 60 40 20 0 Q2 2012 Q3 2012 Q4 2012 Q1 2013 Season Variability Conditions: TN Variability Conditions: AZ Percentage of Days (%) 100 80 60 40 20 0 Q2 2012 Q3 2012 Q4 2012 Q1 2013 Season Percentage of Days (%) 100 80 60 40 20 0 Q2 2012 Q3 2012 Q4 2012 Q1 2013 Season

EPRI High resolution PV Monitoring Concentrated areas include Southeast, Atlantic Coast, and California Clusters PV Plants Installed Recent Install Planned Site Locations, September 2013 Main map: 2012 Microsoft Corporation. Imagery Microsoft available exclusively by DigitalGlobe (Bing Maps Terms of Use: http://go.microsoft.com/?linkid=9710837). Inset map: Map data 2012 Google

Example of Variability on a Feeder in Rome, GA = DPV Site Substation 1 km 2 grid 0.0 mi. 0.5 mi. 1.0 mi. Google Map data Google

Monitoring PV Output Variability F e e d er s Monitoring solar variability and power quality events throughout the feeder Thu Fri Sat 1 2 3 8 9 10 ve Power (kvar) Reactiv 500 400 300 200 100 0-100 -200-300 -400-500 01/13/2013 01/14 01/15 01/16 01/17 01/18 01/19 Using the high resolution data for modeling analysis or Total Tap Counts by Week Regulato 900 800 700 600 500 400 300 200 100 Reg Tap Counts (total by week) Average Variability Index (by week) 0 Apr '12 May '12 Jun '12 Jul '12 Aug '12 Sep '12 Oct '12 Nov '12 Dec '12 Jan '13 Feb '13 12 10 8 6 4 2 0 ility Index (Avg by Week) Variabi

Applying Measured PV Data to Feeder Model Solar Data Collect high res 1 sec solar data to use in model simulations Feeder Modeling Develop feeder models dlwith input from utilities Power Production (W) 1000 Birmingham, AL (May 14, 2010) 800 600 400 200 0 8am 9am 10am 11am 12pm 1pm 2pm 3pm 4pm 5pm Local Time (h) 8976 8977

Analysis Approach Uniqueness from other research efforts Large sample of feeders/characteristics considered 1000 s of potential ti ldeployments (i (size/location) analyzed Range in hosting capacity reported for each issue h Worst-Case Result for Eac Unique PV Deployment 1.07 1.065 1.06 1.055 1.05 1.045 1.04 1.035 Minimum Hosting Capacity Maximum Hosting Capacity A B C Threshold of violation 0 0.5 1 1.5 2 A Allpenetrations in this region are acceptable, regardless of location B Somepenetrations in this region are acceptable, site specific C Nopenetrations in this region are acceptable, regardless of location Details on analysis method: Stochastic Analysis to Determine Feeder Hosting Capacity for Distributed Solar PV. EPRI, Palo Alto, CA: 2012. 1026640. Increasing penetration (MW)

Sample Analysis Spatial &Time Based 3D Visualization of PV Measurement Feeder Voltage Profile PV Production Time Series Regulator/Capacitor Operations

Smart Inverter is a Key Integration Technology DC Power AC Power Traditional Inverter Functionality Matching PV output with grid voltage and frequency Providing safety by providing unintentional islanding protection Disconnect from grid based on over/under voltage/frequency Smart Inverter Functionality Voltage Support Frequency Support Fault Ride Through h (FRT) Communication with grid

What is a Smart Grid Ready Inverter? Link to distributed PV and Storage Systems that become Beneficial distribution system assets Enabling Higher Penetration of Renewables Improving Efficiency Improving Power Quality Reliability Deferred Costs Asset Life

Hosting Capacity with Inverter Support Function PV at Unity Power Factor PV with Volt/var Control Minimum Hosting Capacity Maximum Hosting Capacity Minimum Hosting Capacity Max Hosting Capacity Maximum m Feeder Volt tages (pu) 2500 cases shown Each point = highest primary voltage ANSI voltage limit Maximu um Feeder Vo oltage (pu) ANSI voltage limit Increasing penetration (kw) Increasing penetration (kw) No observable violations regardless of PV size/location Possible violations basedupon PV size/location Observable violations occur regardless of size/location

19 Hosting Impact of Smart Inverter Functions and Set Points Reactive Power Active Power Volt Watt 2 Volt Watt 1 ailable Vars % Av 0.95 100% 1.0V 1.05 % voltage 95%pf 98% pf Volt Var 2 Volt Var 1 Baseline 100% 0 2000 4000 6000 8000 10000 Increasing PV Penetration > % Available Vars Unity Power Factor % voltage All penetrations in this region are acceptable, regardless of location Some penetrations in this region are acceptable, site specific No penetrations in this region are acceptable, regardless of location

Understanding Potential Benefitsof Storage Power Supply Energy and load shifting Regulating services Contingency reserve Power Delivery Asset utilization Investment deferral Power quality and reliability End Use Photovoltaic Energy Shifting Demand Peak Reduction

The Energy Storage Value Concept Balance of Plant Power Electronics Balance of Plant Power Electronics Battery Cost Battery Cost Photovoltaic Peak Shift Demand Peak Reduction Regulation Services Voltage Support Inertia Support Power Quality and Reliability T&D Upgrade Deferral For Illustration Only Customer Side Applications Market Applications Rate Based Applications Costs of Storage Benefits of Storage

The Value of Storage Realities For Illustration Only Balance of Plant Power Balance Electronics of Plant Power Electronics Battery Cost Battery Cost Photovoltaic Peak Shift Demand Peak Reduction Regulation Services Voltage Support Inertia Support Power Quality and Reliability T&D Upgrade Deferral Customer Side Applications Market Applications Rate Based Applications Costs of Storage Benefits of Storage

Various Sources of Flexibility Conventional Resources Peaking and cycling units Emerging Resources Demand response Conventional Gen Energy storage Plug in electric vehicles VG Power Management Control VG output Institution/Market Flexibility Coordination among Balancing Authorities (BAs) Shorter scheduling More Transmission Emerging Resources VG Power Management

Smart Grid Ready Inverter: Objectives 1. Make Inverter Ready and Demo in Field Develop, test, field demo at utility sites Define and incorporate standard grid support functions (500 kva level)* Use feeder/pv modeling for scale up * 2. Revisit Feeder Level Anti Islanding Identify utility controlled trip options Conduct laboratory side by side tests 3. Model DMS DERMS Level Integration Characterize and model support functions Evaluate voltage control strategies and coordination with traditional devices DMS Success depends on utility engagement and site learning

Who are Project Partners? DOE Project EPRI Cost Share SDG&E Xcel Energy Hydro One GRE DTE Energy National Grid AEP Pepco KCP&L Southern Co.

Demonstration Sites 1.7MW (DC) PV 18 95kW Inverters Cedarville, NJ 225kW(DC) PV 300kVA Inverters Ann Arbor, MI 605kW (DC) PV 330kW & 250kW Inverters Everett, MA 1MW (DC) PV 2 500kW Inverters Haverhill, MA

1. Make Inverter Ready Utility resource or customer owned? Markets-drive or grid-code required? Standards and testing? Settings Performance and reliability? What functions and how to initiate them?

Benefits Demonstrated at Different Sites

Benefits Derived from Functions

Testing at ESIF NREL s NRELs Energy System IntegrationFacility

2. Revisit Anti Islanding Enabling Voltage Sag Ride-Through Impacts on DER Hosting Capacity Update standards d for anti-islanding Compare different methods in laboratory Need for feeder level antiislanding solutions

Side by Side comparisons of different technologies Direct transfer trip (DTT) via radio, cell or wire Power line carrier communications (PLCC) low or high frequency Shorting switch Integration of DG into the utility SCADA Synchrophasor based h methods Research Prior Art Develop Functional Requirements Select Suitable Technology for Evaluation Evaluate Technologies in Lab Lb and Field

Power Line Carrier Signal for DER Concept Grid Substation Power Line Tone Generator DX3 Pulsar Transmitter Island Controller Island X X X X

3. Model DMS Level Integration Define functional requirements for DMS-DERMS Address gaps in feeder operational control New Types of DER autonomous or controllable Coordinating with traditional voltage control devices Managing grid control at the edge of the grid

DER Enterprise Integration MDMS OMS GIS DER representation in system model DMS Enterprise Integration DERMS Creation of groups and sharing of group definitions Monitoring of group status Sensors, Switches, Capacitors, Regulators SOLAR BATTERY PEV Dispatch of real and reactive power Forecasting of group capabilities

36 Project Resources https://solarhighpen.energy.gov/segis/electric power research institute i Public Reports: Integrating Smart Distributed Energy Resources with Distribution Management Systems, 1024360 Collaborative Initiative to Advance Enterprise Integration of DER, 1026789 Common Functions for Smart Inverters, Version 2, 1026809 Enterprise Integration Functions for Distributed Energy Resources: Phase 1, 3002001249

Conclusions 1. Smart inverter functions are defined and commercially available. 2. Demonstration in various system feeders, and geographic locations needed. d 3. There are many lessons to learn and share. 4. Challenges: settings, anti-islanding and integration with utility DMS. Thank you! tkey@epri.com