TriForecaster: A Mixture of Experts Framework for Multi-Region Electric Load Forecasting with Tri-dimensional Specialization

2025-08-13

AI Summary

To address the triple challenges of regional heterogeneity, contextual diversity, and temporal dynamics in multi-regional electricity load forecasting (MRELF), this paper proposes TriForecaster—a multi-task learning (MTL)-driven Mixture-of-Experts (MoE) framework. Methodologically, it introduces RegionMixer for adaptive cross-subregion feature fusion and a Context-Time Specializer layer to jointly specialize modeling along contextual and temporal dimensions, enabling dynamic expert scheduling and specialized collaboration. By integrating MoE architecture with MTL, TriForecaster jointly models heterogeneous, multi-source load data. Empirically, it achieves an average 22.4% reduction in forecasting error across four real-world datasets. Deployed on the eForecaster platform in Eastern China, it delivers short-term load forecasts for 17 cities—serving over 110 million residents and supporting over 100 GWh of daily electricity consumption.

Technology Category

• Machine Learning: Mixture of Experts (MoE)
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Electric load forecasting is pivotal for power system operation, planning and decision-making. The rise of smart grids and meters has provided more detailed and high-quality load data at multiple levels of granularity, from home to bus and cities. Motivated by similar patterns of loads across different cities in a province in eastern China, in this paper we focus on the Multi-Region Electric Load Forecasting (MRELF) problem, targeting accurate short-term load forecasting for multiple sub-regions within a large region. We identify three challenges for MRELF, including regional variation, contextual variation, and temporal variation. To address them, we propose TriForecaster, a new framework leveraging the Mixture of Experts (MoE) approach within a Multi-Task Learning (MTL) paradigm to overcome these challenges. TriForecaster features RegionMixer and Context-Time Specializer (CTSpecializer) layers, enabling dynamic cooperation and specialization of expert models across regional, contextual, and temporal dimensions. Based on evaluation on four real-world MRELF datasets with varied granularity, TriForecaster outperforms state-of-the-art models by achieving an average forecast error reduction of 22.4%, thereby demonstrating its flexibility and broad applicability. In particular, the deployment of TriForecaster on the eForecaster platform in eastern China exemplifies its practical utility, effectively providing city-level, short-term load forecasts for 17 cities, supporting a population exceeding 110 million and daily electricity usage over 100 gigawatt-hours.

Problem

• Addresses multi-region electric load forecasting challenges
• Handles regional, contextual, temporal load variations
• Improves accuracy for short-term load predictions

Innovation

• Mixture of Experts for multi-region forecasting
• RegionMixer and CTSpecializer layers for specialization
• Multi-Task Learning to address regional variations