Wind Park Power Prediction: Attention-Based Graph Networks and Deep Learning to Capture Wake Losses

TL;DR

This paper introduces an attention-based graph neural network framework for wind power prediction, effectively modeling wake losses and turbine dependencies, and demonstrating superior performance over traditional models.

Contribution

The paper presents a modular, flexible attention-based GNN architecture that improves wind power prediction and offers interpretability through attention weight analysis.

Findings

Outperforms MLP and BLSTM models in accuracy

Achieves comparable results to vanilla GNNs

Provides insights into turbine dependencies and wake effects

Abstract

With the increased penetration of wind energy into the power grid, it has become increasingly important to be able to predict the expected power production for larger wind farms. Deep learning (DL) models can learn complex patterns in the data and have found wide success in predicting wake losses and expected power production. This paper proposes a modular framework for attention-based graph neural networks (GNN), where attention can be applied to any desired component of a graph block. The results show that the model significantly outperforms a multilayer perceptron (MLP) and a bidirectional LSTM (BLSTM) model, while delivering performance on-par with a vanilla GNN model. Moreover, we argue that the proposed graph attention architecture can easily adapt to different applications by offering flexibility into the desired attention operations to be used, which might depend on the specific application. Through analysis of the attention weights, it was showed that employing attention-based GNNs can provide insights into what the models learn. In particular, the attention networks seemed to realise turbine dependencies that aligned with some physical intuition about wake losses.