Two-Stage Photovoltaic Forecasting: Separating Weather Prediction from Plant-Characteristics

TL;DR

This paper introduces a two-stage photovoltaic forecasting approach that separates weather prediction from plant-specific modeling, analyzing error distributions and improving understanding of forecast inaccuracies for better energy management.

Contribution

It proposes a novel decomposition method for PV forecasting into weather and plant models, utilizing satellite data and neural networks to analyze error sources and distributions.

Findings

Error increases by 11% and 68% when using weather forecasts instead of satellite data.

Generalized hyperbolic and Student's t distributions fit forecast errors well.

Error distribution analysis aids in improving stochastic PV forecasting models.

Abstract

Several energy management applications rely on accurate photovoltaic generation forecasts. Common metrics like mean absolute error or root-mean-square error, omit error-distribution details needed for stochastic optimization. In addition, several approaches use weather forecasts as inputs without analyzing the source of the prediction error. To overcome this gap, we decompose forecasting into a weather forecast model for environmental parameters such as solar irradiance and temperature and a plant characteristic model that captures site-specific parameters like panel orientation, temperature influence, or regular shading. Satellite-based weather observation serves as an intermediate layer. We analyze the error distribution of the high-resolution rapid-refresh numerical weather prediction model that covers the United States as a black-box model for weather forecasting and train an ensemble of neural networks on historical power output data for the plant characteristic model. Results show mean absolute error increases by 11% and 68% for two selected photovoltaic systems when using weather forecasts instead of satellite-based ground-truth weather observations as a perfect forecast. The generalized hyperbolic and Student's t distributions adequately fit the forecast errors across lead times.