Adapting to Change: A Comparison of Continual and Transfer Learning for Modeling Building Thermal Dynamics under Concept Drifts

2025-08-29

AI Summary

In building thermal dynamics modeling, limited data and dynamic environmental changes—such as retrofits or occupancy shifts—induce concept drift, degrading model accuracy over time. To address this, we propose Seasonal Memory Learning (SML), a continual learning strategy that selectively retains historical seasonal knowledge while incrementally integrating new operational data, enabling low-overhead, adaptive model updates. Compared to conventional transfer learning and fine-tuning baselines, SML reduces prediction error by 28.1% under no concept drift and by 34.9% under strong concept drift, outperforming existing approaches. This work introduces, for the first time, a structured memory mechanism into continual learning frameworks for building energy models—achieving a balanced trade-off among predictive accuracy, robustness to distribution shifts, and deployment efficiency. The resulting framework provides a scalable solution for online modeling in dynamically evolving building environments.

Technology Category

• Machine Learning: Life-Long and Continual Learning
• Search and Optimization: Learning to Search
• Cognitive Modeling & Cognitive Systems: Adaptive Behavior

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs

Abstract

Transfer Learning (TL) is currently the most effective approach for modeling building thermal dynamics when only limited data are available. TL uses a pretrained model that is fine-tuned to a specific target building. However, it remains unclear how to proceed after initial fine-tuning, as more operational measurement data are collected over time. This challenge becomes even more complex when the dynamics of the building change, for example, after a retrofit or a change in occupancy. In Machine Learning literature, Continual Learning (CL) methods are used to update models of changing systems. TL approaches can also address this challenge by reusing the pretrained model at each update step and fine-tuning it with new measurement data. A comprehensive study on how to incorporate new measurement data over time to improve prediction accuracy and address the challenges of concept drifts (changes in dynamics) for building thermal dynamics is still missing. Therefore, this study compares several CL and TL strategies, as well as a model trained from scratch, for thermal dynamics modeling during building operation. The methods are evaluated using 5--7 years of simulated data representative of single-family houses in Central Europe, including scenarios with concept drifts from retrofits and changes in occupancy. We propose a CL strategy (Seasonal Memory Learning) that provides greater accuracy improvements than existing CL and TL methods, while maintaining low computational effort. SML outperformed the benchmark of initial fine-tuning by 28.1% without concept drifts and 34.9% with concept drifts.

Problem

• Comparing continual and transfer learning for building thermal modeling
• Addressing concept drifts from retrofits and occupancy changes
• Improving prediction accuracy with new operational measurement data

Innovation

• Proposes Seasonal Memory Learning for thermal modeling
• Compares continual and transfer learning strategies
• Uses simulated long-term data with concept drifts