Prediction of PV Power Production with Neural ODEs on the base of Weather data

Abstract

Predicting energy production from photovoltaics (PV) is crucial for efficient energy management. In order to apply different operating strategies, it is necessary to predict the expected amounts of PV energy. The operating strategies are typically optimized with regard to economic or technical goals or a combination of both. Within this work, we show a possibility to predict PV power production using local weather data and Neural Ordinary Differential Equations (NODE). Based on the measured values from the PV system and an associated weather station, the NODE is trained and validated with regard to PV production. The measurement data are collected from the PV system of the former Campus North of Offenburg University of Applied Sciences.