ASES Resource Application Division Newsletter April 2012

WREF 2012

The World Renewable Energy Network (WREN) and the American Solar Energy Society (ASES) have joined forces to present the World Renewable Energy Forum (WREF). The conference will be held May 13 -17 in Denver, Colorado. Registration is currently open and early bird rates end April 13. Conference details can be found at www.wref2012.org. Sign up now for access to the world’s top renewable energy experts. Researchers from Bangladesh, China, Korea, Iran and the Unites States have submitted abstracts for technical papers to be delivered at the conference.

This international event on renewable energy technologies will address the world’s economic, environmental and security challenges at every scale, from off-grid villages to gigawatt power plants. With over 200 sessions scheduled the conference has something for everybody. Early Bird registration ends April 13. Register at conference/attend/.

WREF 2012 is sponsored by the U.S. Department of Energy’s National Renewable Energy Laboratory (NREL), the Colorado Renewable Energy Society (CRES) and the International Solar Energy Society (ISES).

RAD Members in the News

The current Resource Application Division co-Chair, University of San Diego (UCSD) Assistant Professor and California Solar Collaborative co-Director Dr Jan Kleissl’s research group was recently recognized by President Barack Obama for his Solar Forecasting contributions at the largest PV facility in the United States, Copper Mountain Solar 1. 2012/03/obama-visits-americas-largest-pv-array/

Dr Kleissl and his colleagues at the Jacobs School of Engineering UCSD received a $1.5 million grant from the California Public Utilities Commission (CPUC) earlier this year. The grant funds research to refine forecasting methods and work with San Diego Gas & Electric (SDG&E) to help integrate solar power into the energy grid. Kleissl is the principal investigator on the grant with Carlos Coimbra, also a professor of environmental engineering at UCSD. Their work could generate significant savings for utility companies and increase the wide spread penetration of Solar technology. The two researchers and their teams will work together to develop comprehensive forecast models that predict solar power output, based on the solar resource viewed through a sky imager.

Congratulation Dr Kleissl!

Letter from the Current ASES RAD Chair

I invite you to attend the Resource Application Division meeting during the WREF 2012 conference, Wednesday May 16 at 7:00 AM, breakfast will be served. The RAD group needs your participation more than ever noting it’s time to elect new blood into the Division. The ASES Technical Divisions are one of the core components contributing to the long term success of ASES. It has been a wonderful experience taking a two year ride along with this storied legacy. The once distant phrase grid parity is on the horizon and the solar industries need for accurate solar resource data has never been stronger. The Solar measurement & forecasting knowledge base contained within the Resource Application Division’s 600+ members is second to no other such organization in the United States.

2012 is an election year so the Resource Application Division will be taking nominations for the following positions: Chair, Co-Chair, Secretary, Newsletter Editor, and Technical Reviewer. Please come to the RAD meeting prepared to nominate qualified RAD members with the nominees full name, email, phone number, and employer. If you cannot make the Division meeting during WREF 2012 please email nominations to Justin.robinson@drakerlabs.com by June 15, 2012. Those members eager to step up and take on the challenge make sure your fellow RAD members know you are ready.

I thank all of you who have been involved with the Resource Application Division over the past two years. It is rare in this day and age to find highly successful individuals willing to donate time and effort towards something for no monetary gain. In particular I would like to recognize Jan Kleissl, James Bing, Frank Vignola, and Dave Renne. All of you went above and beyond, thank you for your strong desire to see ASES and the Resource Application Division live on. I commend you, keep up the good work!

I would like to take this opportunity to recognize two Solar Resource Assessment pioneers and wish them the best of luck with retirement. Thank you Daryl Myers and David Renee, NREL will not be the same without you.

Sincerely,

Justin Robinson, ASES RAD Chair Program Manager Data Acquisition, Draker Labs Inc.

Distribution System Scale Solar Monitoring & Forecasting at SMUD

The Sacramento Municipal Utility District (SMUD), the country’s 6th largest publicly owned utility has teamed with NEO Virtus Engineering (NEO), a solar engineering, consulting and monitoring provider, to deploy a service- territory wide solar monitoring network for validating solar forecasting models. Thanks to a grant from the California Public Utilities Commission, under their California Solar Initiative RD&D program, SMUD has initiated a number of projects related to integrating high penetrations of solar PV.

As solar penetrations increase rapidly, SMUD and utilities around the country have begun to focus on questions of how the variability of those systems and the industry’s ability to forecast their output will impact other resources on their system. Forecasting and resource measurement are growing rapidly both within utilities and in the private sector. To better understand the state of forecasting, variability, and future impacts to our system, SMUD and NEO Virtus Engineering have deployed a network of 71 solar monitoring devices covering most of SMUD’s 2330 square kilometer service territory. The project started in June, 2010 and will continue for 2 years. The devices were installed to validate solar forecasts and solar resource variability for high penetrations of solar on SMUD’s grid. Solar monitoring began in May of 2011 and will last for at least 14 months. This study will provide statistical analysis of solar radiation variability across Sacramento County.

REGIONAL IRRADIANCE GROUND TRUTH DATA SET

The purpose of the grant from the CPUC in 2010 was to develop and deploy hardware and software tools to model and mitigate impacts of high penetrations of PV on the distribution network. As part of this research and to validate forecast accuracy, irradiance measurements are being made using a combination of eight RSRs (primary stations) and sixty six global horizontal (GHI) measurements systems (secondary stations). This combination of primary and secondary monitoring stations has been deployed on the same five kilometer square grid as used by the National Digital Forecast Database (NDFD) for their skycover (cloud cover) forecasts. The monitored area spans almost 1775 square kilometers within SMUD’s service territory. All secondary stations take measurements every two seconds and record one minute averages. Data is being retrieved nightly. At each download the logger station clocks are synchronized with the server clock, which is kept at GMT, if difference between logger time and the server is greater than 3 seconds. The monitoring stations have been located in the “nominal centroid” of each 25 square kilometer NDFD cell.

The intent of the data set is to provide body of ground truth data covering a geographic area of a size which could incorporate both distributed generation, of the residential and commercial scale, as well as central PV plant generation of the utility scale, to validate emerging irradiance forecast methodologies. A primary deliverable of this research will be the database of GHI and temperature measurements from the secondary stations and DNI, DHI, GHI and temperature from the primary stations.

IRRADIANCE DATA QUALITY MANAGEMENT

The 5km grid spacing of the measurement network in this project was designed to match the geographic spacing of the National Digital Forecast Database grid. The secondary stations which monitor global horizontal irradiance and ambient temperature have been installed as close as possible to the centroids of the NDFD grid cells. This placement was made possible by using SMUD utility poles as installation locations. With nearly 140,000 distribution poles in the service territory, suitable locations near NDFD grid cell centroids were identified for nearly all grid cells. The population of poles was surveyed to eliminate those with significant shading obstructions and the nearest pole to the centroid was selected. Specific pole selection was done by visually inspecting candidate poles near a centroid to determine whether there were any shading obstructions, and whether there was adequate climbing space on another quadrant of the pole for SMUD linemen. Portions of SMUD’s service territory are in rural areas and in some cases the density of poles did not permit placement close to the centroids. In other cases undergrounded utilities, which now make up nearly 65% of SMUD’s distribution circuit miles meant there were relatively few poles in certain urban areas. In all, 59 of the units are within 500m or less of the centroids. However of the remaining 12 units, five are over 1km from the centroids.

Near Field Shading & Data Filtering

In addition to the issue of proximity to the NDFD grid centroids the experimental design has other practical limitations and compromises. To eliminate shading, mounting the devices on the tops of the distribution poles would have been ideal. However, access to the tops of the poles is difficult and would have required specialized equipment and trained crews, increasing the safety risks and costs of the installation significantly. Instead, as a compromise, the monitoring units were located at the approximate midpoint of the poles. Consequently during portions of the year when the solar zenith angle is relatively small the overhead wires and cross arms briefly cast shadows on the pyranometer. These shadow events present as very discrete and repeatable anomalies on clear sky days, however through the year they move in somewhat difficult to predict patterns depending on the nature of the specific shade element. For instance, some shade elements are distribution wires with differing levels of tautness, where the sun crosses at a different point on the wire each day, creating shifts in time and slight shifts in shade width as the relative distance of the shade object changes throughout the seasons.

Any dips that go below the boundary are identified as shadows. The algorithm counts out the number of shadows that occur on the clear days. Sometimes the program with count a little higher or lower than the correct number shadows, in order to compensate, the algorithm will look at the all the occurring shadows for a station and take the most occurring number of dips for reference. The algorithm then will find the closest clear day to a series of cloudy days. It uses that clear day as a reference point for removing shadows from the cloudy days.

The algorithm uses a similar triangles method to pinpoint the predicted locations of the shadows on cloudy days. Due to the complexity involved with developing a complete 3-D model of each individual site, a simple 2D approach was used. Taking conservative approach, three minutes worth of data was removed from each side of the identified shadow to make sure that the complete shadow was removed. A resulting edit of a slightly cloudy day can be viewed in Figure 6. Notice that the gaps are larger than in the previous clear day because of the buffering. SMUD will continue to develop algorithms to minimize the amount of lost data associated with these shade objects, however for this analysis, the small buffer approach is adequate.

Sensor Error and System Wide Calibration

All of the pyranometers used in this project were Licor 200SZ devices. All were new when deployed and their 2 year calibration spanned the term of the project. However a limitation of the experimental design is the inability to clean the pyranometers once they are installed on the poles. To address this issue four Eppley Precision Spectral Pyranometers (PSP) were installed on four of the RSR primary stations. The four PSPs are located approximately in the four compass points of the map of the monitored area.

Prior to installation all four of the PSPs were calibrated in the National Renewable Energy Laboratory’s (NREL) Broadband Outdoor Radiometer Calibration (BORCAL) process1.

Before publishing the complete dataset we will develop a correction procedure using the PSP data. The procedure will:

• Locate a very clear sky day in the data record
• Based upon BORCAL information correct known measurement errors inherent in the PSPs
• Average the BORCAL corrected clear sky measurements from the four PSPs and take the averaged value as our reference
• Compare the PSP average to each of the 66 GHI values at solar noon and several other points in the day and derive a scale factor which will bring the GHI into agreement with the corrected PSPs
• “Adjust” the data set for each GHI unit using the scale factors until the next clear sky day which follows the next scheduled cleaning of the PSPs
• Provide the raw secondary station GHI data, the raw PSP data as well as the BORCAL PSP corrections

CONCLUSION

The deployment of 71 irradiance sensors covering SMUD’s service territory has allowed a unique opportunity for researchers to validate a variety of solar forecasting approaches over a previously unavailable large geographic area. The pole-mounted sensors allowed for deployment in a grid-like pattern to closely align with the NDFD weather forecast grid. Further, they provided a secure location for the devices to operate autonomously with remote cell-modem based data collection. However, they also presented unique challenges in that there are shade objects present at each location resulting from distribution wires, cross-arms, and trees or buildings. Two methods for eliminating these shade objects were developed, and further work in refinement of automated methods continue to ensure the maximum amount of this data is usable. These data sets, both primary and secondary, in both raw and filtered forms, will be made available for download to the research community in the near future.

AUTHORS

James Bing, NEO Virtus Engineering Inc., jbing@neovirtus.com; Obadiah Bartholomy and Thomas Vargas, SMUD, Obadiah.Bartholomy@smud.org, Thomas.Vargas@smud.org; Pramod Krishnani, Belectric Inc., krishnani.pramod@gmail.com, pramod.krishnani@belectric-usa.com

REFERENCES

(1) Wilcox, S.M, Myers, D.R. “Evaluation of Radiometers in Full-Time Use at the National Renewable Energy Laboratory Solar Radiation Research Laboratory” December 2008, NREL NREL/TP-550-44627

Solar & Infrared Radiation Measurements

A new solar instrumentation text book “Solar & Infrared Radiation Measurements” is coming out at the end of June that culminates several years of effort by Joe Michalsky, Tom Stoffel, and Frank Vignola. Each of the authors has been involved with solar monitoring for over thirty years and has had much experience and many publications to draw upon. Ashley Gasque (ashley.gasque@taylorandfrancis.com) of CRC press asked the group to put together a solar instrumentation book after Dr Vignola presented a paper on pyrheliometers at the International Society for Optics and Photonics (SPIE) conference in San Diego.

The book is written for both students and professionals and includes an introduction to solar measurement fundamentals. General background and terminology needed to understand the uses and requirements of various solar monitoring radiometers is provided along with an overview on the various solar radiation instruments. The main focus of the book is global, diffuse, and direct normal shortwave radiation measurements but also contains chapters on infrared, net, UV, and spectral measurements. The book also includes sections on meteorological measurements, a step by step guide to setting up a solar monitoring station, and appendices on modeling solar data and how to use satellite images to estimate solar radiation. It also discusses the calibration and maintenance of instruments in the field, as well as covering the traceability of calibrations to international standards. For more information, go to http://www.crcpress.com/ and type the title of the book in the search window.

WREF 2012 Content of Interest (RAD)

Solar Resource Measurement (Technical), 5/14/2012 10:30am – 11:45am

• Reducing Uncertainty in Bankable Solar Resource and Energy Assessments through On-Site Monitoring
• Solar Monitoring, Forecasting, and Variability Assessment at SMUD
• Calibration of Long-Term Global Horizontal Irradiation Estimated By HelioClim-3 through Short-Term Local Measurement
• Characterizing Measurement Campaigns for the Calibration of the Diffuse Horizontal Irradiation Estimated By HelioClim-3
• Can Silicon Detectors be used for Measuring PV Performance?

Resource Variability and Modeling (Technical), 5/14/2012 1:15pm-2:30pm

• Impact of Distributed Generation on PV Variability: A Lanai Case Study
• The Variability Index: A New and Novel Metric for Quantifying Irradiance and PV Output Variability
• Optimum Fixed Orientations Considering Day-Ahead Market Energy Pricing in California
• GHI Correlations With DHI And DNI And The Effects Of Cloudiness On One-Minute Data
• Development of a Stochastic-Kinematic Cloud Model to Generate High-Frequency Solar Irradiance and Power Data
• A High-Resolution Surface Radiation Dataset from Geostationary Satellites: Methodology and Validation

Solar Resource Forecasting-1 (Technical), 5/14/2012 2:45pm – 4:00pm

• Design and Performance of an Optimized Ensemble Solar Generation Forecast System for Grid Systems on the Hawaiian Islands
• Cloud Advection Schemes for Short-Term Satellite-Based Insolation Forecasts
• Validation and Analysis of HRRR Insolation Forecasts using SURFRAD
• Critical Timeframes of Importance for PV from a Utility Perspective
• ECMWF Forecast Assessment of Direct Solar Irradiance over Australia
• An International Solar Irradiance Data Ingest System For Forecasting Solar Power And Agricultural Crop Yields

Solar Forecasting: Needs and Solutions (Forum), 5/14/2012 4:15pm – 5:30pm

• FORUM – Solar Forecasting: Needs and Solutions
• FORUM – Solar Forecasting for Integration of Large Amounts of Solar Energy

Renewable Energy Resource Assessment Methods & Applications Poster Session Exhibit Hall D5/15/2012 10:30am – 11:45am

• High Resolution Solar Measurements at DeSoto: The Deployment Experience
• Study of Solar Energy Potential in Azerbaijan
• Model To Calculate PV Array Altitude And Azimuth Angles To Maximize Energy And Demand Revenues From Measured Hourly Solar Radiation And Building Use Data
• An Artificial Neural Network Based Approach for Estimating Direct Normal, Diffuse Horizontal and Global Horizontal Irradiances using Satellite Images
• Recalibration of Heliosat-2 Method for Global Horizontal Irradiance Estimation in Dusty and Humid Environments
• Quantifying the Use of Site-Specific Measurements in Place of a Typical Meteorological Year Data Set for Solar Energy Modeling
• Improvement of the Weather Research and Forecasting (WRF) Mesoscale Model for Improved Solar Resource Assessments and Forecasts under clear skies
• Forecasting Solar Power Intermittency using Ground-Based Cloud Imaging
• Comparing Two Linear Regression Techniques for Long-Term Wind Speed Forecasting
• Assessment of Wind Energy Resources in Algeria
• Wind Properties at Turbine Hub Height: Examples Using Doppler Lidar
• Cost effective Wave measurements for Ocean energy
• Solar resource assessment in coastal NW-Europe
• Modeled Yearly Energy Yield Of Inverted Metamorphic Multijunction Solar Cells

Solar Resource Forecasting-II (Technical) 5/15/2012 10:30am – 11:45am

• Towards Intra-Hour Solar Forecasting Using Two Sky Imagers At A Large Solar Power Plant
• Determination of Forecast Value Considering Energy Pricing in California
• Short Term DNI Forecasting With Sky Imaging Techniques
• Cloud velocity estimation from an array of solar radiation measurements
• Development and testing of a new day-ahead solar power forecasting system
• Characterization of Irradiance Variability Using a High-resolution, Cloud Assimilating NWP

Resource Assessment Methods (Ignite) 5/15/2012 1:15pm – 2:30pm

• Atmospheric Attenuation of Solar Radiation in Central Receiver Systems
• Lidar measurements for offshore Wind Energy research
• P50/P90 Analysis for Solar Energy Systems Using the System Advisor Model
• System Advisor Model (SAM) Case Studies Comparing To Real Performance Results
• Analysis on the Mathematical Modeling of Wind Speed Probability Distribution Function
• Climate-Regime Cospectrum Analysis: Shortwave Solar Irradiance for Regionally Spaced Locales
• Computing Solar Energy Potential of Urban Areas Using Airborne LIDAR and Orthoimagery
• Analyzing Temporal and Spatial Variations of Direct Normal, Diffuse Horizontal and Global Horizontal Irradiances Estimated from an Artificial Neural Network Based Model
• Evaluation of Procedures to Improve Solar Resource Assessments: Optimum Use of Short-Term Data from a Local Weather Station to Correct Bias in Long-Term Satellite-Derived Solar Radiation Time Series
• Reanalysis: An Improved Data Set For Simulating Wind Generation?
• A Review of Solar Resource Assessment Initiatives in South Africa: The Case for a National Network
• Wind Speed Profile: A new Artificial Neural Network–Power Law Model

WREN-IV – Solar Radiation and Associated Topics (Technical), 5/15/2012 1:15pm – 2:30pm

• Energy Efficiency and Renewable Energy – A Path to Sustainability in Buildings
• POTENTIALS FOR SOLAR IRRADIANCE ESTIMATION FROM TEMPERATURE DATA FOR AKOKA – NIGERIA
• A Bayesian Committee Model Approach To Forecasting Global Solar Radiation
• SOLAR IRRADIATION FORECASTING: STATE-OF-THE-ART AND PROPOSITION FOR FUTURE DEVELOPMENTS FOR SMALL-SCALE INSULAR GRIDS
• Comparison of Clear-Sky and Persistence Models for Evaluating Solar Forecasting Skill
• Measuring Irradiance, Temperature and Angle of Incidence Effects on Photovoltaic Modules Using a Sourcemeter-based Test-bed
• Conformity of Ground Truth with Satellite Derived Data: The Ilorin Experience of Downward Longwave Radiation

Solar Resource Methods (Technical), 5/15/2012 2:45pm – 4:00pm

• Optimising The Temporal Averaging Period Of Point Surface Solar Resource Measurements For Correlation With Areal Satellite Estimates
• Improving Modeled Solar Irradiance Historical Time Series: What is the Appropriate Monthly Statistics for Aerosol Optical Depth?
• Understanding the Variation in Estimated Long-Term Solar Resource Estimates; Which Data Set Accurately Represents Your Project Site?
• Quantifying the Accuracy of the Use of Measure-Correlate-Predict Methodology for Long-Term Solar Resource Assessments
• Reporting of Irradiance Model Relative Errors

Solar Variability & Forecasting (Ignite), 5/15/2012 4:15pm – 5:30pm

• High-Density Solar Measurements at DeSoto: A Step Towards Developing a Better Understanding of PV Variability
• Predicting Short-Term Variability of High-Penetration PV
• A Simple Cloud Simulator for Investigating the Correlation Scaling Coefficient Used in the Wavelet Variability Model (WVM)
• A Comparison of Wind Power and Load Forecasting Error Distributions
• Outputs and error indicators for solar forecasting models
• Increasing time resolution of satellite-derived solar irradiance time-series
• PROGRESS TOWARD A VIRTUAL PYRANOMETER FOR LOW COST PERFORMANCE MONITORING OF PV SYSTEMS
• Modeling Distribution System Impacts of Solar Variability and Interconnection Location -Quantifying Long-Timescale Solar Resource Variability

Solar Resources and Forecasting Workshop Highlights (Forum), 5/16/2012 10:30am – 11:45am

• FORUM – Solar Resources & Forecasting Workshop Highlights

Photovoltaics & Wind Poster Session Exhibit Hall D, 5/16/2012 10:30am – 11:45am

• Spectral Distributions of Diffuse and Global Irradiance for Clear and Cloudy Periods

Resource Applications (Technical), 5/17/2012 10:30am – 11:45am

• A Consensus Wind forecasting System
• Clean Energy Ministerial Global Atlas for Solar and Wind Energy
• Micro-scale Numerical Weather Prediction and Its Application for an Offshore Wind Farm in South Korea
• Towards a High Resolution Long-term Solar Resource Database: Applying the SUNY model to ISCCP B1U data stream

From Minimizing Risk to Maximizing Performance: Flat Panel PV Resource Assessment Best Practices (Forum) , 5/17/2012 4:15pm – 5:30pm

• FORUM: From Minimizing Risk to Maximizing Performance: Flat Panel PV Resource Assessment Best Practices
• FORUM: ISO-9060 Standard & Pyranometer Measurement Accuracy
• FORUM: Flat Panel PV Resource Assessment Qualitative Analysis: Reference Cell or Pyranometer

Tour of the National Renewable Energy Laboratory (NREL), Golden, CO – ticket required, 5/18/2012 9:00am – Noon

Tour of NREL’s National Wind Technology Center, Boulder, CO – ticket required, 5/18/2012 2:00pm – 4:00pm