Satellite-Based Software Tools for Optimizing Utility Planning, Simulation and Forecasting Tom Hoff, President, Research & Consulting ISES Webinar February 23, 2015 Copyright 2015 Clean Power Research, L.L.C
Webinar Agenda Introduction to Clean Power Research The distributed PV energy challenge Solutions based on satellite data technology Applications that support solar utilities 2
SOLAR PREDICTION Founded in 1998 with the mission to power intelligent energy decisions Most widely used solar resource database PROGRAM OPTIMIZATION Research Consulting Software ~4.7 GW of renewable incentives processed ENERGY VALUATION ~30 million solar estimations performed 3
Complete Solutions by Market Segment Solar Industry Lead Generation & Quoting Incentive & Interconnection Processing Performance Benchmarking for Fleet O&M PowerBill APIs PowerClerk API SolarAnywhere SystemCheck API Utility / ISO / Energy Agency Ratepayer Engagement Incentive & Interconnection Processing Grid Planning & Operations WattPlan PowerClerk SolarAnywhere FleetView
The PV Challenge: Plan and Forecast Load was separated from generation
The PV Challenge: Plan and Forecast Now, generation is co-located with load
The PV Challenge: DG Capacity is Growing (and fast)!
MW Changing Load Shapes 2007-2015 14000 13000 12000 11000 10000 9000 2/7/2007 1/14/2008 2/19/2009 1/14/2010 1/27/2011 2/6/2012 1/9/2013 1/27/2014 1/22/2015 8000 7000 6000 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 Data courtesy of the California Independent System Operator (CAISO) 8
Management of Distributed Solar Administration Incentives Interconnection Customer Engagement Education Intelligence Grid Integration Grid Design & Planning Grid Operations
Management of Distributed Solar Administration System of Record Customer Engagement Grid Integration
Output Calculations Data FleetView PV Simulation Methodology PV Specifications PowerClerk Solar Irradiance Data SolarAnywhere Satellite & NWP PV Simulation Engine PV Production Forecasts FleetView methodology also supports historical simulation needs 11
How satellite measurement works Clear sky radiation NOAA GOES satellite images Haze (turbidity) Clouds Reflectivity (albedo)
Hybrid, multi-model forecasting Ramp and Near-Term Satellite-derived cloud motion vector (CMV) Sub-hourly forecasts in the near-term Probabilistic ranges Medium-Term Blended Numerical Weather Prediction Models (NWP) Day-ahead forecasts up to seven days ahead Probabilistic ranges 12 hours Real ahead time Location of Interest
SolarAnywhere Data Web-accessible solar irradiance data Irradiance data Historical satellite-derived timeseries data from 1998 through latest hour 1-10 km spatial and hourly, halfhourly and one-minute temporal resolution options available Forecasts up to 7-days in advance by combining cloud motion vector and NWP approaches 14
Three SolarAnywhere Resolutions Standard Resolution 10 km, 1 hour Enhanced Resolution 1 km, ½ hour High Resolution 1 km, 1 minute 10 km 1 km 1 km Example: San Francisco, CA
Explicit BTM PV Fleet Simulation Individually consider all PV systems on the grid Use high spatially- and temporally-synchronized weather and irradiance prediction (accurately represent spatial correlation between PV sites) Directly (physically) model PV output FleetView offers accurate, scalable PV simulation solutions 16
Captures BTM Fleet Locations and Diversity PowerClerk links administration to fleet simulation Same methodology as employed with utility-scale PV simulations Schneider Electric Inverter (GT-100-208-PG) 352 SunPower 327 W SPR-327NE-WHT-D 10 Tilt, 181 Azimuth 33.705 N, -117.876 W Commissioned Jan. 2014 Regional and system-wide BTM simulation capabilities 17
Advantages of Explicit PV Fleet Simulations Example of DG solar installation diversity Significant variance in PV energy production by orientation Wide distribution of PV systems can alter fleet PV energy output 18
Validation in Load Forecasting Models Itron and CAISO are evaluating CPR s BTM PV fleet historical production as training input into CAISO s ALFS PG&E sub-region was modeled from Jan 2010 through Feb 2014 PG&E sub-region 19
Preliminary Results (Historical Training) Hour of Day Ahead Forecast 9 am 12 pm 3 pm CPR BTM Dataset? No Yes No Yes No Yes Load Forecast Error (MW) 122.3 121.7 136.7 130.1 149.3 142.7 Load Forecast Error (%) 1.11% 1.10% 1.18% 1.12% 1.27% 1.21% BTM Coefficient - -0.21 - -0.92 - -0.94 T-test (significant if < - 1.64) - -1.8 - -10.1 - -8.9 A BTM coefficient of -1 is the ideal result (i.e., for every predicted MW of BTM generation, one MW of load is shed) BTM production datasets offer a statistically significant improvement for mid-day and late-afternoon load forecasting results lie quite close to the theoretical value of -1 BTM production datasets offer a smaller impact for morning load forecasting Believed to result from higher morning load variability 20
California Metered PV System Simulations Accurate utility-scale PV simulations require detailed site specifications (PV modules, inverters, orientation, row spacing, etc.). Same methodology as pre-construction PV energy simulations (PVsyst, SAM, etc.) Aggregated CAISO Fleet PV production 120+ Metered Utility-Scale PV Systems in the CAISO region 21
PV Fleet Power Forecasting Note: Utility Sited systems include intertie systems in NV and AZ
Demand (MW) Resource Planning Hourly synchronized PV production and utility-load 60,000 50,000 1.7%/yr. growth 40,000 30,000 20,000 10,000 2020 Peak (w/12 GW PV) 2012 Peak (w/1.3 GW PV) System Demand Total Demand 0 0:00 6:00 12:00 18:00 0:00 Time of Day Peak Day: August 13, 2012
Resource Planning DEF ramp rates, non-residential fleet, 2013 +/- 1.3% of fleet rating 24
Distribution Engineering Interconnection studies PV Database Circuit Model 29942037 210 kw,... 29942038 450 kw,... 29942039 10 kw,... 29942040 75 kw,... 29942041 850 kw,... 29942042 20 kw,... Solar Data PV Analysis 25
Distribution Operations Smart grid Substation A Substation B N/O GROUP A PV PV N/C PV PV PV Voltage Regulator PV GROUP A.2 GROUP A.1 26
Active Research Continuing on this Topic DOE SUNRISE Demonstrate improved net utility load forecasts by incorporating behind-the-meter PV forecasts for CAISO and all PV in California EPIC (CPR Teamed with Itron) Address cost-effective strategies for integrating large amounts of PV into distribution systems by integrating PV modeling into utility planning and operation tools 27
Thank you Please feel free to contact us for any details or clarification related to presentation Skip Dise SolarAnywhere Prod. Manager johndise@cleanpower.com Adam Kankiewicz Solar Research Scientist adamk@cleanpower.com Grant Brohm Senior Account Executive grantb@cleanpower.com The information herein is for informational purposes only and represents the current view of Clean Power Research, L.L.C. as of the date of this presentation. Because Clean Power Research must respond to changing market conditions, it should not be interpreted to be a commitment on the part of Clean Power Research, and Clean Power Research cannot guarantee the accuracy of any information provided after the date of this presentation. CLEAN POWER RESEARCH, L.L.C. MAKES NO WARRANTIES, EXPRESS, IMPLIED OR STATUTORY, AS TO THE INFORMATION IN THIS PRESENTATION.
Scalable PV simulation solutions Step 1: Build PV Fleet from and Other Sources Step 2: Obtain Solar Resource Data from Forecast Historical Step 3: Simulate PV Fleet Production Using FleetView System Operation Capacity Planning
Identifying the PV Generators San Francisco PG&E Bay Area PG&E Non Bay Area SCE Coastal SCE Inland SDG&E Detailed Individual PV system specs (location, modules, inverter, layout, etc. details) 30