Trends and gaps in photovoltaic power forecasting with machine learning
A. Alcañiz (TU Delft - Photovoltaic Materials and Devices)
D. Grzebyk (Student TU Delft)
H. Ziar (TU Delft - Photovoltaic Materials and Devices)
O. Isabella (TU Delft - Photovoltaic Materials and Devices)
More Info
expand_more
Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.
Abstract
The share of solar energy in the electricity mix increases year after year. Knowing the production of photovoltaic (PV) power at each instant of time is crucial for its integration into the grid. However, due to meteorological phenomena, PV power output can be uncertain and continuously varying, which complicates yield prediction. In recent years, machine learning (ML) techniques have entered the world of PV power forecasting to help increase the accuracy of predictions. Researchers have seen great potential in this approach, creating a vast literature on the topic. This paper intends to identify the most popular approaches and the gaps in this discipline. To do so, a representative part of the literature consisting of 100 publications is classified based on different aspects such as ML family, location of PV systems, number of systems considered, features, etc. Via this classification, the main trends and gaps can be highlighted while offering advice to researchers interested in the topic.
help_outline