PV output variability modeling using satellite imagery and neural networks

Abstract

High frequency irradiance variability measured on the ground is caused by the formation, dissipation, and passage of clouds in the sky. Variability and ramp rates of PV systems are increasingly important to understand and model for grid stability as PV penetration levels rise. Using satellite imagery to identify cloud types and patterns can predict irradiance variability in areas lacking sensors. With satellite imagery covering the entire U.S., this allows for more accurate integration planning and power flow modelling over wide areas. Satellite imagery from southern Nevada was analyzed at 15 minute intervals over a year. Methods for image stabilization, cloud detection, and textural classification of clouds were developed and tested. High Performance Computing parallel processing algorithms were also investigated and tested. Artificial Neural Networks using imagery as inputs were trained on ground-based measurements of irradiance to model the variability and were tested to show some promise as a means for predicting irradiance variability. Artificial Neural Networks, cloud texture analysis, and cloud type categorization can be used to model the irradiance and variability for a location at a one minute resolution without needing many ground based irradiance sensors.