Metamodel to predict annual cooling thermal load for commercial, services and public buildings: A country-level approach to support energy efficiency regulation
ENERGY AND BUILDINGS(2023)
摘要
The energy sector significantly impacts the environment, with energy production contributing to greenhouse gases emissions and climate change. In Brazil, buildings account for a substantial portion of energy consumption, making energy efficiency essential for sustainable development. Building simulation is an efficient way to provide valuable insights into the thermal performance of buildings, but it requires expertise, time, and computational resources. To overcome these simulation constraints, metamodeling has emerged as an easy-to-use and fast-response alternative to analyse the thermal performance of buildings. This study focuses on developing a metamodel to predict cooling thermal loads in Brazil's commercial, services, and public buildings, supporting the country's building energy efficiency labelling program. It is expected from the metamodel a high capacity to reproduce the variability of climates, contexts, and heterogeneity of buildings from a country-level perspective. A parametric sampling process was used to develop a comprehensive simulated database considering several variations of building-related, occupancy patterns, and weather parameters. The metamodel was trained, validated and tested using optimisation techniques and an artificial neural network. Afterwards, it was compared with actual models, considering different typologies and climates. While the metamodel demonstrates high accuracy and generalisation, limitations were found regarding its application in warmer temperatures and complex building shapes. Further refinement is needed to improve its applicability and reliability in real-world scenarios. The proposed metamodel offers a practical and widely applicable tool for supporting energy code compliance and energy efficiency assessment in buildings.
更多查看译文
关键词
Metamodel,Building energy performance,Building energy simulation,Brazilian regulation
AI 理解论文
溯源树
样例
![](https://originalfileserver.aminer.cn/sys/aminer/pubs/mrt_preview.jpeg)
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要