Short-term Prediction and Filtering of Solar Power Using State-Space Gaussian Processes

TL;DR

This paper develops a scalable Gaussian process model using state-space methods and variational inference for accurate short-term solar power prediction with uncertainty estimates, suitable for large datasets and real-time streaming.

Contribution

It introduces a novel combination of state-space Gaussian processes and variational inference to enable scalable, real-time solar power forecasting with uncertainty quantification.

Findings

Model handles large PV datasets efficiently

Provides accurate short-term predictions with error bars

Capable of real-time streaming data filtering

Abstract

Short-term forecasting of solar photovoltaic energy (PV) production is important for powerplant management. Ideally these forecasts are equipped with error bars, so that downstream decisions can account for uncertainty. To produce predictions with error bars in this setting, we consider Gaussian processes (GPs) for modelling and predicting solar photovoltaic energy production in the UK. A standard application of GP regression on the PV timeseries data is infeasible due to the large data size and non-Gaussianity of PV readings. However, this is made possible by leveraging recent advances in scalable GP inference, in particular, by using the state-space form of GPs, combined with modern variational inference techniques. The resulting model is not only scalable to large datasets but can also handle continuous data streams via Kalman filtering.