Long-term Wind Power Forecasting with Hierarchical Spatial-Temporal Transformer

TL;DR

This paper introduces HSTTN, a hierarchical spatial-temporal Transformer model for long-term wind power forecasting, effectively capturing multi-scale dependencies and spatial-temporal correlations to improve prediction accuracy.

Contribution

The paper proposes a novel end-to-end hierarchical Transformer framework with a unique hourglass architecture and contextual fusion for enhanced long-term wind power prediction.

Findings

HSTTN outperforms existing methods on real-world datasets.

The model effectively captures long-range temporal dependencies.

Spatial-temporal feature fusion improves forecasting accuracy.

Abstract

Wind power is attracting increasing attention around the world due to its renewable, pollution-free, and other advantages. However, safely and stably integrating the high permeability intermittent power energy into electric power systems remains challenging. Accurate wind power forecasting (WPF) can effectively reduce power fluctuations in power system operations. Existing methods are mainly designed for short-term predictions and lack effective spatial-temporal feature augmentation. In this work, we propose a novel end-to-end wind power forecasting model named Hierarchical Spatial-Temporal Transformer Network (HSTTN) to address the long-term WPF problems. Specifically, we construct an hourglass-shaped encoder-decoder framework with skip-connections to jointly model representations aggregated in hierarchical temporal scales, which benefits long-term forecasting. Based on this framework, we capture the inter-scale long-range temporal dependencies and global spatial correlations with two parallel Transformer skeletons and strengthen the intra-scale connections with downsampling and upsampling operations. Moreover, the complementary information from spatial and temporal features is fused and propagated in each other via Contextual Fusion Blocks (CFBs) to promote the prediction further. Extensive experimental results on two large-scale real-world datasets demonstrate the superior performance of our HSTTN over existing solutions.