Neural network interpretability for forecasting of aggregated renewable generation
This addresses the need for reliable forecasting in renewable energy management for decision-makers, but it is incremental as it applies existing interpretability and uncertainty methods to a specific domain.
The paper tackles the problem of forecasting aggregated renewable generation and whether solar power exceeds load for small prosumers, using interpretable neural networks with gradient-based methods and uncertainty estimation to provide robust and explainable predictions.
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 gradient-based methods and complemented with uncertainty estimation, provide robust and explainable forecasting for decision-makers.