# Comparative review of hybrid forecasting models for short-term prediction of building thermal load

Published: 2026-10-07 · Updated: 2026-10-07 · Category: Research

Source page: https://www.sipply.dev/story/comparative-review-of-hybrid-forecasting-models-for-short-443501cf · 中文: https://www.sipply.dev/zh/story/comparative-review-of-hybrid-forecasting-models-for-short-443501cf

## Summary

A recent paper has conducted a comparative review of hybrid forecasting models aimed at short-term prediction of building thermal loads. The study highlights the performance and methodologies of various models, ultimately identifying the EMD-LSTM-Markov model as the most accurate for day-ahead predictions of heating and domestic hot water demands. However, it notes that this model faces challenges when dealing with high power swells, indicating areas for improvement in its application. The review serves as a valuable resource for researchers and practitioners looking to enhance energy efficiency in building management through better forecasting techniques.

## Key takeaways

- The EMD-LSTM-Markov model is the most accurate for day-ahead thermal load predictions.
- Hybrid models show promise in improving forecasting accuracy for building energy demands.
- Challenges remain in handling extreme power conditions, highlighting the need for further research.

## Why it matters

Improving forecasting models for building thermal loads can lead to significant energy savings and enhanced efficiency in building management, contributing to sustainability efforts.

- **Product managers:** Product managers should consider integrating advanced forecasting models into their energy management solutions to provide clients with more accurate predictions and optimize energy usage.
- **Developers:** Developers can leverage insights from this research to enhance existing forecasting algorithms or create new models that address the limitations identified, particularly in handling high power swells.
- **Students and job seekers:** Students and job-seekers in the fields of energy management and data science can benefit from understanding these hybrid models, as they represent a growing area of research with practical applications in sustainability.

## Sources

- [Comparative review of hybrid forecasting models for short-term prediction of building thermal load](https://arxiv.org/abs/2610.06881) (arXiv cs.LG, 2026-10-07)

## 中文

### 建筑热负荷短期预测的混合预测模型比较评审

最近的一篇论文对旨在短期预测建筑热负荷的混合预测模型进行了比较评审。研究强调了各种模型的性能和方法，最终确定EMD-LSTM-Markov模型为最准确的日预测供暖和生活热水需求的模型。然而，它指出该模型在处理高功率波动时面临挑战，表明其应用中存在改进空间。该评审为希望通过更好的预测技术提高建筑管理能效的研究人员和从业者提供了宝贵的资源。

**要点：**

- EMD-LSTM-Markov模型是日预测热负荷的最准确模型。
- 混合模型在提高建筑能源需求预测准确性方面显示出潜力。
- 在处理极端功率条件时仍然存在挑战，强调了进一步研究的必要性。

**为什么重要：** 改善建筑热负荷的预测模型可以带来显著的能源节约和提高建筑管理的效率，促进可持续发展努力。

---
Sipply: Sip AI News Simply. https://www.sipply.dev