SolarAnywhere Real-time Model Validation

Updated August 2023
SolarAnywhere Support Center

Executive Summary

SolarAnywhere® SystemCheck® is a software-as-a-service product maintained by Clean Power Research®, providing on-demand access to accurate, real-time solar irradiance, weather and production data for the trailing 45 days up to the current hour, globally. The software and data quality are mature in development and adoption.

SystemCheck is widely used for operational PV system monitoring and performance benchmarking. Real-time solar irradiance and weather data are used to independently estimate weather-normalized PV system production for direct comparison to actual system production. SolarAnywhere irradiance data are generated using a satellite-based model that uses visible- and infrared-channel data from a global network of geosynchronous orbiting satellites. Resulting data are spatially and temporally consistent, and geographically precise.

SolarAnywhere real-time irradiance data are generated using the same steps and model input fields as SolarAnywhere historical data, but with several variations in input data sources due to availability and optimizations to enable faster performance.

SolarAnywhere real-time global horizontal irradiance (GHI) generally performs within 1% of SolarAnywhere historical model Version 3.7 (V3.7) in North America. Figure 1 summarizes annual, monthly and hourly relative mean absolute error (rMAE) and relative root mean square error (rRMSE) for the reference year 2022. These statistics are calculated for the SolarAnywhere real-time data model and historical model V3.7 across a trusted network of well-maintained ground measurement stations in the United States. They are representative of product performance, but should not be taken as an absolute indicator of accuracy given the uncertainty introduced into the study by the reference dataset. For additional information, see Validation Methodology.

Figure 1: SolarAnywhere Real-time GHI Accuracy Metrics

1 Difference = Absolute Value (Historical – Real-Time)
2 Annual rMAE and rRMSE are identical because only one year is considered in the calculation


This document provides up-to-date reference information and validation statistics for Clean Power Research’s SolarAnywhere® SystemCheck® real-time irradiance data product. The document version 2023.09 was last updated in September 2023.

About Clean Power Research

Clean Power Research® has delivered award-winning cloud software solutions to utilities and industry for more than 20 years. Our PowerClerk®, WattPlan® and SolarAnywhere® product families allow our customers to make sense of, and thrive amid, the energy transformation. Clean Power Research has offices in Napa, Calif., and Bellevue, Wash. For more information, visit

About SolarAnywhere

​SolarAnywhere irradiance data are generated using visible- and infrared-channel data captured by geosynchronous orbiting satellites. The satellite images are processed using the most advanced algorithms developed by Dr. Richard Perez at the University at Albany (SUNY). These algorithms extract cloud indices from the satellite’s visible and infrared data. A self-calibrating feedback process adjusts for arbitrary ground surfaces such as terrain and albedo. The cloud indices are used to modulate physically based radiative transfer models describing localized clear sky climatology.

The Perez model is applied in a pseudo-empirical fashion that is periodically calibrated with a select few ground stations. However, it operates largely independent of ongoing ground data input. This approach is unique to the industry and enables ground-to-satellite correlation studies to be truly based on two independently derived measurement sources.

SolarAnywhere irradiance data are generated in both global horizontal (GHI) and direct normal (DNI) irradiance components. The following geometric balancing equation is used to calculate diffuse horizontal irradiance (DHI):

DHI = GHI - DNI\times cos(α\times zenith)

Clean Power Research has an exclusive relationship with Dr. Perez and SUNY to implement the latest satellite-to-solar irradiance methodology advances. More information on the extensive validation of the Perez model can be found in the references section.

In agreement with the U.S. Department of Energy through the National Renewable Energy Laboratory (NREL), Perez model-based satellite irradiance data comprised the 2005 (SUNY version 1) and 2010 (SolarAnywhere version 2.3) National Solar Radiation Database (NSRDB) releases. While the output format of SolarAnywhere satellite irradiance data is similar to NSRDB data, SolarAnywhere now provides more recent and more accurate datasets intended for commercial use.

The newest version of the SolarAnywhere model has been implemented operationally as Version 3.7 (V3.7). SolarAnywhere satellite irradiance data are available for specific sites on a 1 km x 15-min or 10 km x hourly basis, and from 1998 to the present hour depending on geographic availability. High resolution, 0.5 km x 5-min data are are available for the continental United States beginning Jan. 1, 2020.

Validation Methodology

Data from select ground stations are used as a reference to calculate the error and uncertainty of the operational SolarAnywhere real-time irradiance model. A high-quality reference dataset is required for the validation statistics to represent, to the maximum extent possible, model performance rather than ground data inaccuracies. Therefore, only the highest quality reference stations are used as validation sites, and the data are screened for data quality issues. The validation sites span a wide geographic area and a variety of terrain and climate types to assess the model’s performance in heterogeneous conditions. Most are part of the World Radiation Monitoring Center Baseline Surface Radiation Network (BSRN).

The SolarAnywhere model has three critical properties for the purposes of data validation and overall confidence in the model’s performance. First, the model is never fit to individual validation sites. Second, the model operates independently of ongoing ground data input. Third, SolarAnywhere utilizes a single real-time model regardless of the time and location (with adaptations for each satellite platform). Because of these properties, the validation statistics are representative of the model’s performance not just at the validation sites, but also for the model generally.

The GHI, DNI and Diffuse Horizontal Irradiance (DIF) data are compared at hourly, daily, monthly and annual intervals using traditional error metrices such as relative Mean Absolute Error (MAE) and Root Mean Square Error (RMSE).

The error metrics are defined by the following formulas:

  1. rMAE=\frac{\sum_{i=1}^N | (x_i^{SA} - x_i^{obs}) | }{N}\frac{100\%}{\overline{x_i^{obs}}}
  2. rRMSE=\sqrt\frac{\sum_{i=1}^N (x_i^{SA} - x_i^{obs})^{2}}{N}\frac{100\%}{\overline{x_i^{obs}}}

An 𝑥 represents the variable being considered (either GHI, DNI or DHI); 𝑁 is the number of data points used; and the superscripts SA and obs stand for SolarAnywhere and ground observed data. The error metrics are normalized by the mean of the ground observed data and denoted by 𝑟MAE and 𝑟RMSE. Hoff et al. have previously discussed the applicability of various error metrics for solar in the paper Reporting of Irradiance Model Relative Errors.”

The statistics presented here are representative of product performance, but should not be taken as an absolute indicator of accuracy. Despite best efforts to quality control the reference data, no reference dataset is perfect. Where possible, SolarAnywhere data is tested against GHI calculated from measured direct and diffuse irradiance (DHI + DNI*cos(αzenith)). The component sum measurements are generally more accurate than pyranometer measurements of GHI; however, such measurements are not available for all validation sites. For consistency, the GHI statistics presented in this document use pyranometer measurements as the reference.

New real-time data values are generated every 30 minutes based on the latest satellite image. The real-time model is built to provide the best-available irradiance and weather estimates for any given time and location. This means that until data is re-generated using the operational historical data models and archived, real-time data values are subject to change based on the time of the request.

Visit our support center to learn more about the transition between SolarAnywhere real-time and historical data. Real-time data values used for this validation were generated assuming no satellite image retrieval delays, which can occur in the operational model on rare occasion (<0.5% of the time in 2022). While SolarAnywhere offers real-time data at 1 km nominal resolution, real-time data for this validation was generated at 10 km nominal resolution. It is expected that higher resolution (1 km nominal) real-time data will demonstrate increased accuracy.

Validation Results

Asset managers may use monthly satellite-based PV production data to understand PV performance in the context of recent weather. 

Since satellite-based PV production data is weather normalized, it can be used to isolate actual plant performance from variability in monthly insolation, temperature, etc. Hourly error metrics are useful for understanding how the model performs in various weather conditions. In general, as the averaging period is shortened (e.g., from annual to monthly, or monthly to hourly), errors increase due to the fundamental properties of averages. 

Figure 2 summarizes monthly and hourly error for the SolarAnywhere real-time model and historical model V3.7 for the reference year 2022, as well as historical model V3.7 for the reference years 1998-2022.

Figure 2: SolarAnywhere Real-time and Historical Annual, Monthly and Hourly Error

SolarAnywhere real-time global horizontal irradiance (GHI) generally performs within 1% of SolarAnywhere historical model V3.7 in North America for the reference year 2022. The real-time model for the reference year 2022 also demonstrates good alignment with the historical model over the full available period (1998 – 2022), where rMAE values are within 1% and rRMSE values are within 1.1%.

The GHI relative mean absolute error for each validation site is plotted in the maps in Figure 3 below. Each point represents the average monthly rMAE for the reference year 2022.

Figure 3: SolarAnywhere Real-time GHI Monthly rMAE by Validation Site

For all locations, the rMAE of GHI falls between 0% and 3.9% for 11 out of the 12 sites (excluding Madison, Wisconsin). The Madison, Wisconsin site demonstrates a monthly rMAE of 9.0%. The Madison site presents complex modeling circumstances due to its close proximity to Lake Michigan.

Figure 4: Half-hourly Real-time GHI (W/m^2)

Desert Rock, 2022

Penn State, 2022


The SolarAnywhere real-time model demonstrates performance consistent with that of the historical model. Annual, monthly and hourly error calculated for the SolarAnywhere real-time model falls within 1% of equivalent statistics calculated for the historical model V3.7. Monthly and hourly statistics demonstrate the model’s ability to capture shorter periods and the range of possible weather for operational use cases.

The real-time model is also spatially consistent. Monthly rMAE falls between 0% and 3.9% across 11 geographically dispersed validation sites. Scatterplots of real-time data against ground measurements demonstrate good model performance in both low and high irradiance conditions. SolarAnywhere ensures the highest data quality globally by using a single, satellite-to-solar model everywhere. Future iterations of this study will expand the validation into additional geographic regions.

Clean Power Research continues to invest in SolarAnywhere to meet the needs of the solar industry and accelerate the clean energy transformation.


Input Data Sources

SolarAnywhere uses a satellite-to-solar algorithm to estimate irradiance from geosynchronous satellite images. The raw input data for the irradiance data include:

Auxiliary data including aerosol optical depth, wind speed, ambient temperature, relative humidity, solid precipitation, liquid precipitation and snow depth are derived from various numerical weather models. 

This validation of the SolarAnywhere real-time model is limited to the North America satellite region. Future iterations of this study will expand to additional geographic regions. 

Reference Stations

Meaningful validation requires a high quality refence. To ensure the validation statistics are representative of the model, the validation stations are required to meet the following criteria:

  • A credible organization maintains responsibility for the installation
  • Metadata such as sensor type, location, etc. exists; sensors are secondary standard or better
  • The data are publicly available
  • The period of record is at least 1 year
  • The data generally pass standard quality control and the station has >75% availability
  • The station is representative of locations where solar PV is installed

The following stations were used in this SolarAnywhere real-time data validation study. They met the criteria above and had data available for the reference year 2022.

Quality Control

Data from each station must pass statistical data quality checks similar to those recommended by the BSRN. In addition, data are reviewed by a data analyst to ensure the measurements are not affected by the following common issues:

  • Soiling 
  • Shading 
  • Calibration drift

Data that fails to pass the quality control are excluded from the analysis. If the issues with the station are persistent, the station is excluded from the analysis entirely. Deviation from SolarAnywhere is not used as a reason to exclude data. 


Detailed statistics by site are available in the print version of this document.


The following peer-reviewed articles describe the SUNY model underlying SolarAnywhere simulations:
  1. Perez R., P. Ineichen, K. Moore, M. Kmiecik, C. Chain, R. George and F. Vignola, (2002): A New Operational Satellite-to-Irradiance Model. Solar Energy 73, 5, p. 307-317.
  2. Perez R., P. Ineichen, M. Kmiecik, K. Moore, R. George and D. Renné, (2004): Producing satellite-derived irradiances in complex arid terrain. Solar Energy 77, 4, p. 363-370.
  3. Perez, R., P. Ineichen, E. Maxwell, R. Seals and A. Zelenka, (1992): Dynamic Global-to-Direct Irradiance Conversion Models. ASHRAE Transactions-Research Series, p. 354-369.
  4. P. Ineichen, (2008): Comparison and validation of three global-to-beam irradiance models against ground measurements. Solar Energy 82, p. 501-512.
Reference stations reference:
  1. BSRN: Driemel, A., Augustine, J., Behrens, K., Colle, S., Cox, C., Cuevas-Agulló, E., Denn, F. M., Duprat, T., Fukuda, M., Grobe, H., Haeffelin, M., Hodges, G., Hyett, N., Ijima, O., Kallis, A., Knap, W., Kustov, V., Long, C. N., Longenecker, D., Lupi, A., Maturilli, M., Mimouni, M., Ntsangwane, L., Ogihara, H., Olano, X., Olefs, M., Omori, M., Passamani, L., Pereira, E. B., Schmithüsen, H., Schumacher, S., Sieger, R., Tamlyn, J., Vogt, R., Vuilleumier, L., Xia, X., Ohmura, A., and König-Langlo, G.: Baseline Surface Radiation Network (BSRN): structure and data description (1992– 2017), Earth Syst. Sci. Data, 10, 1491-1501, doi:10.5194/essd-10-1491-2018, 2018.
Need a printable version of this validation document?

To see all SolarAnywhere validation documents, click the link below.