Spatiotemporal Graph Neural Networks in short term load forecasting: Does adding Graph Structure in Consumption Data Improve Predictions?

2025-02-14

AI Summary

This paper addresses the challenge of modeling implicit spatial structure in smart meter data for short-term load forecasting (STLF), where no explicit graph topology is available. Methodologically, it systematically evaluates spatiotemporal graph neural networks (STGNNs) without prior graph assumptions by integrating graph structure learning, multi-scale temporal convolution, and graph attention mechanisms, while comparing correlation- and geography-based graph construction strategies to capture fine-grained spatiotemporal dependencies in residential electricity consumption. The key contribution is the first empirical identification of granularity-dependent graph modeling gains: STGNNs yield an average 8.2% reduction in MAPE for residential-level forecasting—demonstrating substantial performance improvement—but exhibit negligible gains at aggregated (e.g., feeder or substation) levels. This finding clarifies the applicability boundary of graph neural networks in load forecasting and provides both theoretical insight and empirical evidence to guide the principled deployment of STGNNs in practical STLF scenarios.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Data Mining & Knowledge Management: Mining of Spatial, Temporal or Spatio-Temporal Data

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Semantics and Knowledge: Scalable techniques for the creation, curation, publication, maintenance, and consumption of large, Web-based, structured, reusable, knowledge graphs and ontologies

Abstract

Short term Load Forecasting (STLF) plays an important role in traditional and modern power systems. Most STLF models predominantly exploit temporal dependencies from historical data to predict future consumption. Nowadays, with the widespread deployment of smart meters, their data can contain spatiotemporal dependencies. In particular, their consumption data is not only correlated to historical values but also to the values of neighboring smart meters. This new characteristic motivates researchers to explore and experiment with new models that can effectively integrate spatiotemporal interrelations to increase forecasting performance. Spatiotemporal Graph Neural Networks (STGNNs) can leverage such interrelations by modeling relationships between smart meters as a graph and using these relationships as additional features to predict future energy consumption. While extensively studied in other spatiotemporal forecasting domains such as traffic, environments, or renewable energy generation, their application to load forecasting remains relatively unexplored, particularly in scenarios where the graph structure is not inherently available. This paper overviews the current literature focusing on STGNNs with application in STLF. Additionally, from a technical perspective, it also benchmarks selected STGNN models for STLF at the residential and aggregate levels. The results indicate that incorporating graph features can improve forecasting accuracy at the residential level; however, this effect is not reflected at the aggregate level

Problem

• Explores STGNNs for short-term load forecasting
• Assesses graph structure impact on forecasting accuracy
• Benchmarks STGNN models at different consumption levels

Innovation

• Spatiotemporal Graph Neural Networks
• Graph Structure Integration
• Residential Level Accuracy Improvement