A Deep Learning Framework for Heat Demand Forecasting using Time-Frequency Representations of Decomposed Features

2026-03-01

AI Summary

This study addresses the challenge of multi-step heat load forecasting in district heating systems, which is complicated by nonlinear dynamics and external factors. The authors propose a novel deep learning framework that integrates continuous wavelet transform (CWT) with time–frequency feature modeling. By combining the time–frequency representations derived from CWT with convolutional neural networks (CNNs), the method effectively captures peak fluctuations in heat demand and uncovers underlying patterns in key features. Extensive validation on multi-year datasets from three Danish regions and two cities in Germany demonstrates that the model reduces mean absolute error by 36%–43% compared to the strongest baseline, achieving an annual prediction accuracy of 95%. This performance significantly surpasses that of conventional statistical models, Transformers, and emerging foundation models.

Technology Category

• Machine Learning: Time-Series/Data Streams
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Search and Optimization: Learning to Search

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

District Heating Systems are essential infrastructure for delivering heat to consumers across a geographic region sustainably, yet efficient management relies on optimizing diverse energy sources, such as wood, gas, electricity, and solar, in response to fluctuating demand. Aligning supply with demand is critical not only for ensuring reliable heat distribution but also for minimizing carbon emissions and extending infrastructure lifespan through lower operating temperatures. However, accurate multi-step forecasting to support these goals remains challenging due to complex, non-linear usage patterns and external dependencies. In this work, we propose a novel deep learning framework for day-ahead heat demand prediction that leverages time-frequency representations of historical data. By applying Continuous Wavelet Transform to decomposed demand and external meteorological factors, our approach enables Convolutional Neural Networks to learn hierarchical temporal features that are often inaccessible to standard time domain models. We systematically evaluate this method against statistical baselines, state-of-the-art Transformers, and emerging foundation models using multi-year data from three distinct Danish districts, a Danish city, and a German city. The results show a significant advancement, reducing the Mean Absolute Error by 36% to 43% compared to the strongest baselines, achieving forecasting accuracy of up to 95% across annual test datasets. Qualitative and statistical analyses further confirm the accuracy and robustness by reliably tracking volatile demand peaks where others fail. This work contributes both a high-performance forecasting architecture and critical insights into optimal feature composition, offering a validated solution for modern energy applications.

Problem

• heat demand forecasting
• district heating systems
• multi-step forecasting
• non-linear patterns
• external dependencies

Innovation

• time-frequency representation
• continuous wavelet transform
• heat demand forecasting
• deep learning
• convolutional neural networks