STDHL: Spatio-Temporal Dynamic Hypergraph Learning for Wind Power Forecasting

TL;DR

This paper introduces STDHL, a novel spatio-temporal hypergraph learning model that captures complex, dynamic correlations among wind farms to improve ultra-short-term wind power forecasting accuracy.

Contribution

The paper proposes a new hypergraph-based deep learning model with dynamic hypergraph convolution and grouped temporal layers for better modeling of wind farm correlations.

Findings

Outperforms existing state-of-the-art methods on GEFCom dataset.

Highlights the importance of spatio-temporal covariates in forecasting accuracy.

Demonstrates the effectiveness of hypergraph structures in capturing higher-order spatial features.

Abstract

Leveraging spatio-temporal correlations among wind farms can significantly enhance the accuracy of ultra-short-term wind power forecasting. However, the complex and dynamic nature of these correlations presents significant modeling challenges. To address this, we propose a spatio-temporal dynamic hypergraph learning (STDHL) model. This model uses a hypergraph structure to represent spatial features among wind farms. Unlike traditional graph structures, which only capture pair-wise node features, hypergraphs create hyperedges connecting multiple nodes, enabling the representation and transmission of higher-order spatial features. The STDHL model incorporates a novel dynamic hypergraph convolutional layer to model dynamic spatial correlations and a grouped temporal convolutional layer for channel-independent temporal modeling. The model uses spatio-temporal encoders to extract features from multi-source covariates, which are mapped to quantile results through a forecast decoder. Experimental results using the GEFCom dataset show that the STDHL model outperforms existing state-of-the-art methods. Furthermore, an in-depth analysis highlights the critical role of spatio-temporal covariates in improving ultra-short-term forecasting accuracy.