Interpretable Load Forecasting via Representation Learning of Geo-distributed Meteorological Factors

2025-01-04

AI Summary

To address the challenges of dynamic selection and low interpretability in fusing multi-source geographical and meteorological factors (MFs) for short-term power load forecasting, this paper proposes a graph-based meteorological factor representation learning framework. We construct a regional meteorology–load heterogeneous graph that jointly encodes spatial correlations and economic–industrial priors. To quantify each MF’s regional contribution to load, we introduce Shapley value attribution and design a Monte Carlo acceleration algorithm for efficient computation. Finally, interpretable predictions are achieved via weighted linear regression. Evaluated on real-world datasets, our method significantly improves 1–3-day-ahead forecasting accuracy—especially under extreme conditions such as summer heat accumulation and winter temperature shocks. Empirical analysis further reveals strong statistical correlations between MF importance rankings and regional GDP levels as well as dominant industrial sectors. To the best of our knowledge, this is the first work to systematically integrate graph neural networks, game-theoretic attribution, and geospatial representation learning into load forecasting.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Knowledge Representation and Reasoning: Preferences
• Reasoning under Uncertainty: Graphical Models

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Meteorological factors (MF) are crucial in day-ahead load forecasting as they significantly influence the electricity consumption behaviors of consumers. Numerous studies have incorporated MF into the load forecasting model to achieve higher accuracy. Selecting MF from one representative location or the averaged MF as the inputs of the forecasting model is a common practice. However, the difference in MF collected in various locations within a region may be significant, which poses a challenge in selecting the appropriate MF from numerous locations. A representation learning framework is proposed to extract geo-distributed MF while considering their spatial relationships. In addition, this paper employs the Shapley value in the graph-based model to reveal connections between MF collected in different locations and loads. To reduce the computational complexity of calculating the Shapley value, an acceleration method is adopted based on Monte Carlo sampling and weighted linear regression. Experiments on two real-world datasets demonstrate that the proposed method improves the day-ahead forecasting accuracy, especially in extreme scenarios such as the"accumulation temperature effect"in summer and"sudden temperature change"in winter. We also find a significant correlation between the importance of MF in different locations and the corresponding area's GDP and mainstay industry.

Problem

• Weather Variability
• Electricity Demand Forecasting
• Regional Economic Impact

Innovation

• spatially integrated meteorological data
• Shapley value analysis
• enhanced demand forecasting accuracy