Conference paper
Neural network interpretability for forecasting of aggregated renewable generation
Technical University of Denmark1
Department of Electrical Engineering, Technical University of Denmark2
Center for Electric Power and Energy, Centers, Technical University of Denmark3
Electric Power Systems, Center for Electric Power and Energy, Centers, Technical University of Denmark4
Energy Analytics and Markets, Center for Electric Power and Energy, Centers, Technical University of Denmark5
With the rapid growth of renewable energy, lots of small photovoltaic (PV) prosumers emerge. Due to the uncertainty of solar power generation, there is a need for aggregated prosumers to predict solar power generation and whether solar power generation will be larger than load. This paper presents two interpretable neural networks to solve the problem: one binary classification neural network and one regression neural network.
The neural networks are built using TensorFlow. The global feature importance and local feature contributions are examined by three gradient-based methods: Integrated Gradients, Expected Gradients, and DeepLIFT. Moreover, we detect abnormal cases when predictions might fail by estimating the prediction uncertainty using Bayesian neural networks.
Neural networks, which are interpreted by the gradient-based methods and complemented with uncertainty estimation, provide robust and explainable forecasting for decision-makers.
Language: | English |
---|---|
Publisher: | IEEE |
Year: | 2021 |
Pages: | 282-288 |
Proceedings: | 2021 IEEE International Conference on Communications, Control, and Computing Technologies for Smart Grids |
ISBN: | 1665415029 and 9781665415026 |
Types: | Conference paper |
DOI: | 10.1109/SmartGridComm51999.2021.9631993 |
ORCIDs: | Murzakhanov, Ilgiz and Chatzivasileiadis, Spyros |