Physics-Informed Neural Networks for predicting indoor temperature and cooling demand in historic buildings

作者
Simona Semeraro,Francesca Vecchi,Roberto Stasi,Umberto Berardi
出处
期刊:Journal of building engineering [Elsevier BV]
卷期号:115: 114392-114392 被引量:19
标识
DOI:10.1016/j.jobe.2025.114392
摘要

Buildings are complex assets, characterised by evolving uses, variable occupancies, and long-life cycles, which lead to high operational costs and significant energy demands. Globally, buildings are responsible for 30–40% of greenhouse gas emissions, making accurate energy forecasting essential for reducing both economic and environmental impacts. Despite the growing adoption of Machine Learning (ML) methods, these approaches often require large datasets and struggle to maintain physical consistency, particularly in data-scarce contexts such as historic buildings, where energy efficiency measures must also comply with heritage constraints. To address these challenges, this study proposes a hybrid predictive framework based on Physics-Informed Neural Networks (PINNs), dynamic energy simulations, and energy balance constraints to forecast, one hour ahead, the indoor operative temperature (T o ) and cooling electricity consumption (E el,Cool ) of a historic public building in southern Italy. The PINN operates in two stages: first predicting T o,t+1 and then estimating E el,Cool,t+1 by combining data-driven learning with thermodynamic principles. This approach reduces reliance on computationally expensive simulations, shortens prediction times, and provides physically consistent predictions. Results confirm the superiority of the PINN over conventional ML models, achieving an RMSE of 0.091 °C (CVRMSE = 0.34%, MBE = 0.025 °C, and NMBE = 0.09%) for T o and 2.12 kWh (CVRMSE = 9.95%, MBE = 0.04 kWh, and NMBE = 0.19%) for E el,Cool . Compared to MLP, Random Forest, and Linear Regression, the PINN reduced RMSE by 96.8%, 96.7%, and 98.6% respectively for T o , and by 26.1%, 24.3%, and 68.4% for E el,Cool . These findings highlight the potential of PINNs to bridge the gap between data scarcity and physical interpretability, enabling robust energy forecasting and comfort optimization in heritage contexts. • Physics-Informed Neural Networks and Bayesian hyperparameter optimization are combined • A two-phase PINN architecture is used to estimate operative temperature and cooling demand • The model was trained using simulated hourly data for the summer season • PINN achieved superior predictive accuracy compared to other machine learning models • The potential of PINNs for accurate and interpretable modelling in data-scarce cases is confirmed

科研通智能强力驱动
Strongly Powered by AbleSci AI
科研通是完全免费的文献互助平台,具备全网最快的应助速度,最高的求助完成率。 对每一个文献求助,科研通都将尽心尽力,给求助人一个满意的交代。
实时播报
徐文欣完成签到 ,获得积分10
刚刚
1秒前
qian发布了新的文献求助10
1秒前
DKO253完成签到,获得积分10
1秒前
1秒前
椒盐天蛾完成签到 ,获得积分10
1秒前
顾矜的应助被cbx采纳,获得10
2秒前
2秒前
Hase发布了新的文献求助30
2秒前
2秒前
2秒前
傻死一只橙子完成签到,获得积分10
3秒前
yimuyi发布了新的文献求助10
3秒前
qqqqq完成签到,获得积分10
3秒前
4秒前
江江发布了新的文献求助20
4秒前
4秒前
ada发布了新的文献求助10
4秒前
4秒前
4秒前
AAA卫生院保洁杨姐完成签到 ,获得积分10
4秒前
好好活着的应助被孤独梦曼采纳,获得10
5秒前
5秒前
星辰大海的应助被hhy采纳,获得10
6秒前
碳水诈蛋完成签到 ,获得积分10
6秒前
简单的白云完成签到,获得积分10
6秒前
归海听荷发布了新的文献求助10
6秒前
莉莉安发布了新的文献求助10
6秒前
7秒前
01发布了新的文献求助10
7秒前
咕咕day发布了新的文献求助10
8秒前
丫丫发布了新的文献求助10
8秒前
CipherSage的应助被强风吹拂采纳,获得10
8秒前
双shuang发布了新的文献求助10
8秒前
lalala发布了新的文献求助10
8秒前
刘宇翔发布了新的文献求助10
8秒前
anian完成签到,获得积分10
8秒前
eclo完成签到 ,获得积分10
9秒前
wanci的应助被繁星采纳,获得10
9秒前
9秒前
高分求助中
(应助此贴封号)【重要!!请各用户(尤其是新用户)详细阅读】【科研通的精品贴汇总】 10000
拉伸高分子熔体中缠结行为的模拟和小角中子散射研究 1500
The World Atlas of Wine 9th edition 1000
奈特基础组织学:配套组织病理学(第4版)(Netter's Essential Histology With Correlated Histopathology) 800
John McDowells Bildungstheorie 800
八旗制度の研究 600
Analyses de reconnaissances de dettes passées devant les échevins d'Ypres (1249-1291) 600
热门求助领域 (近24小时)
化学 材料科学 医学 生物 纳米技术 计算机科学 工程类 有机化学 化学工程 生物化学 物理 复合材料 内科学 催化作用 光电子学 人工智能 细胞生物学 无机化学 物理化学 数学
热门帖子
关注 科研通微信公众号,转发送积分 8160632
求助须知:如何正确求助?哪些是违规求助? 9555512
关于积分的说明 21728801
捐赠科研通 7750444
什么是DOI,文献DOI怎么找? 3427429
关于科研通互助平台的介绍 2552456
邀请新用户注册赠送积分活动 2471817