Operational day-ahead solar power forecasting for aggregated PV systems with a varying spatial distribution

Lennard Visser,Tarek Alskaif,Wilfried Van Sark

doi:10.1016/j.renene.2021.10.102

Lennard Visser, Tarek Alskaif + Show 1 more

Open Access

https://doi.org/10.1016/j.renene.2021.10.102

Copy DOI

Abstract

Accurate forecasts of the power production of distributed photovoltaic (PV) systems are essential to support grid operation and enable a high PV penetration rate in the electricity grid. In this study, we analyse the performance of 12 different models that forecast the day-ahead power production in agreement with market conditions. These models include regression, support vector regression, ensemble learning, deep learning and physical based techniques. In addition, we examine the effect of aggregating multiple PV systems with a varying inter-system distance on the forecast model performance. The models are evaluated both on their technical and economic performance. From a technical perspective, the results show a positive effect from both an increasing inter-system distance and a larger sized PV fleet on the model performance, which was not the case for the economic assessment. Furthermore, the ensemble and deep learning models perform better than the alternatives from a technical point of view. For the economic assessment, the results indicate the superiority of the physical based model, followed by the deep learning models. Lastly, our findings show the importance of considering the user's objective when assessing solar power forecast models.

Highlights

Vast decreasing costs associated to solar photovoltaic (PV) systems have increased the competitiveness of PV systems to other power generation technologies
The performance of the forecast models is examined to their technical (MAE) and economic (ER) performance, where the latter concerns a novel application to evaluating solar forecasting methods
The results show that from a technical perspective and in case of forecasting the PV power production for a single PV system, the best performance is obtained by the RF and subsequently the Long Short-Term Memory (LSTM) model

Summary

Introduction

Vast decreasing costs associated to solar photovoltaic (PV) systems have increased the competitiveness of PV systems to other power generation technologies This has resulted in a surge of the global installed capacity of PV systems in recent years. Reliable forecasts of PV power output have the potential to support stable grid operation when the PV penetration rate increases, while limiting the need for balancing reserves. These forecasts are in the first place valuable for grid operators as it reduces the integration costs associated with PV systems [6]. Accurate forecasts are identified as a requirement to facilitate a high PV penetration level [8]

Objectives

Results

Conclusion