Globalization for Scalable Short-term Load Forecasting

2025-07-15

AI Summary

Traditional local forecasting models (LFMs) suffer from poor generalizability, overfitting, sensitivity to data drift and cold-start conditions, and limited scalability in power transmission networks. To address these limitations, this paper proposes a scalable global short-term load forecasting framework. Methodologically, it integrates time-series clustering with weighted instance learning to jointly model global feature transformation and target transformation, enabling cross-regional collaborative learning. The key contributions are: (i) the first systematic characterization of how global modeling enhances peak-load and hierarchical forecasting performance, and (ii) quantitative evaluation of the adaptability differences among various transformation models under heterogeneous data conditions. Extensive experiments on real-world power load data demonstrate that the proposed model significantly outperforms local baselines—achieving a 12.6% improvement in prediction accuracy under large-scale heterogeneous scenarios, while simultaneously enhancing robustness and generalization capability.

Technology Category

• Machine Learning: Scalability of ML Systems
• Search and Optimization: Local Search
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems

Application Category

• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization

Abstract

Forecasting load in power transmission networks is essential across various hierarchical levels, from the system level down to individual points of delivery (PoD). While intuitive and locally accurate, traditional local forecasting models (LFMs) face significant limitations, particularly in handling generalizability, overfitting, data drift, and the cold start problem. These methods also struggle with scalability, becoming computationally expensive and less efficient as the network's size and data volume grow. In contrast, global forecasting models (GFMs) offer a new approach to enhance prediction generalizability, scalability, accuracy, and robustness through globalization and cross-learning. This paper investigates global load forecasting in the presence of data drifts, highlighting the impact of different modeling techniques and data heterogeneity. We explore feature-transforming and target-transforming models, demonstrating how globalization, data heterogeneity, and data drift affect each differently. In addition, we examine the role of globalization in peak load forecasting and its potential for hierarchical forecasting. To address data heterogeneity and the balance between globality and locality, we propose separate time series clustering (TSC) methods, introducing model-based TSC for feature-transforming models and new weighted instance-based TSC for target-transforming models. Through extensive experiments on a real-world dataset of Alberta's electricity load, we demonstrate that global target-transforming models consistently outperform their local counterparts, especially when enriched with global features and clustering techniques. In contrast, global feature-transforming models face challenges in balancing local and global dynamics, often requiring TSC to manage data heterogeneity effectively.

Problem

• Addressing limitations of local load forecasting models in scalability and accuracy
• Exploring global forecasting models to improve prediction robustness and generalizability
• Proposing clustering methods to manage data heterogeneity in load forecasting

Innovation

• Global forecasting models enhance scalability and accuracy
• Time series clustering addresses data heterogeneity effectively
• Target-transforming models outperform local models with clustering