Neural network interpretability for forecasting of aggregated renewable generation

TL;DR

This paper introduces interpretable neural networks for forecasting aggregated renewable solar power generation, utilizing gradient-based methods and uncertainty estimation to enhance robustness and explainability for decision-makers.

Contribution

The paper develops two neural networks with interpretability and uncertainty estimation, advancing explainable forecasting of renewable energy with gradient-based explanations and Bayesian methods.

Findings

Neural networks accurately predict solar power generation.

Gradient-based methods effectively interpret feature importance.

Uncertainty estimation identifies potential prediction failures.

Abstract

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.