Journal article
Increasing the Accuracy of Hourly Multi-Output Solar Power Forecast with Physics-Informed Machine Learning
Department of Electrical Engineering, Technical University of Denmark1
Dynamical Systems, Department of Applied Mathematics and Computer Science, Technical University of Denmark2
Center for Electric Power and Energy, Centers, Technical University of Denmark3
Energy System Management, Center for Electric Power and Energy, Centers, Technical University of Denmark4
Department of Photonics Engineering, Technical University of Denmark5
Diode Lasers and LED Systems, Department of Photonics Engineering, Technical University of Denmark6
Photovoltaic Materials and Systems, Department of Photonics Engineering, Technical University of Denmark7
Department of Wind Energy, Technical University of Denmark8
GRID Integration and Energy Systems, Wind Energy Systems Division, Department of Wind Energy, Technical University of Denmark9
Department of Applied Mathematics and Computer Science, Technical University of Denmark10
...and 0 moreMachine Learning (ML)-based methods have been identified as capable of providing up to one day ahead Photovoltaic (PV) power forecasts. In this research, we introduce a generic physical model of a PV system into ML predictors to forecast from one to three days ahead. The only requirement is a basic dataset including power, wind speed and air temperature measurements.
Then, these are recombined into physics informed metrics able to capture the operational point of the PV. In this way, the models learn about the physical relationships of the different features, effectively easing training. In order to generalise the results, we also present a methodology evaluating this physics informed approach.
We present a study-case of a PV system in Denmark to validate our claims by extensively evaluating five different ML methods: Random Forest, Support Vector Machine, Convolutional Neural Networks (CNN), Long-Short Term Memory (LSTM) and a hybrid CNN–LSTM. The results show consistently how the best predictors use the proposed physics-informed features disregarding the particular ML-method, and forecasting horizon.
However, also, how there is a threshold regarding the number of previous samples to be included that appears as a convex function.
Language: | English |
---|---|
Publisher: | MDPI |
Year: | 2022 |
Pages: | 749 |
ISSN: | 14243210 and 14248220 |
Types: | Journal article |
DOI: | 10.3390/s22030749 |
ORCIDs: | Pombo, Daniel Vázquez , Bindner, Henrik W. , Spataru, Sergiu Viorel , Sørensen, Poul Ejnar and Bacher, Peder |