Explainable Load Forecasting with Covariate-Informed Time Series Foundation Models

2026-04-30

AI Summary

This work addresses the challenge of interpretability in time series foundation models for electric load forecasting, where opacity hinders adoption in safety-critical infrastructure. The study introduces an efficient SHAP-based explanation method tailored to time series foundation models such as Chronos-2 and TabPFN-TS, employing a novel joint masking strategy for temporal and covariate features. This approach enables scalable feature attribution while maintaining high-accuracy zero-shot forecasting performance. Evaluated on day-ahead load forecasting tasks for transmission system operators, the model matches the predictive accuracy of specialized Transformers trained over multiple years, while its explanations align closely with domain knowledge—particularly regarding weather and calendar effects—thereby substantially enhancing the trustworthiness and practical utility of the forecasting system.

Abstract

Time Series Foundation Models (TSFMs) have recently emerged as general-purpose forecasting models and show considerable potential for applications in energy systems. However, applications in critical infrastructure like power grids require transparency to ensure trust and reliability and cannot rely on pure black-box models. To enhance the transparency of TSFMs, we propose an efficient algorithm for computing Shapley Additive Explanations (SHAP) tailored to these models. The proposed approach leverages the flexibility of TSFMs with respect to input context length and provided covariates. This property enables efficient temporal and covariate masking (selectively withholding inputs), allowing for a scalable explanation of model predictions using SHAP. We evaluate two TSFMs - Chronos-2 and TabPFN-TS - on a day-ahead load forecasting task for a transmission system operator (TSO). In a zero-shot setting, both models achieve predictive performance competitive with a Transformer model trained specifically on multiple years of TSO data. The explanations obtained through our proposed approach align with established domain knowledge, particularly as the TSFMs appropriately use weather and calendar information for load prediction. Overall, we demonstrate that TSFMs can serve as transparent and reliable tools for operational energy forecasting.

Problem

• Explainable AI
• Load Forecasting
• Time Series Foundation Models
• Transparency
• Critical Infrastructure

Innovation

• Time Series Foundation Models
• Explainable AI
• SHAP
• Load Forecasting
• Covariate-Informed