Building and Environment最新文献_第6页

Human metabolic pollutants in bedrooms at typical occupancy density: Composition analysis and effects on sleep quality 典型占用密度下卧室中人体代谢污染物的成分分析及其对睡眠质量的影响

IF 7.6 1区工程技术 Q1 CONSTRUCTION & BUILDING TECHNOLOGY

Building and Environment

Pub Date : 2026-01-14 DOI: 10.1016/j.buildenv.2026.114247

Xiaojing Zhang , Yiting Liu , Mei Wang , Jingchao Xie

The formulation of effective air quality and ventilation standards for sleep environment depends on fundamental data of human emission characteristics and a clear understanding of their impact on sleep quality. This study quantified CO₂ and VOC concentrations in thermally neutral confined bedrooms under three typical occupant densities, monitoring sleep stages, environmental perceptions, and physiological responses in eight healthy adults. Results showed an inverse relationship between per-capita CO₂ emission rate and CO₂ exposure concentration. Relative to the baseline concentration (1,063 ± 182 ppm), this relationship necessitated correction factors of 0.79 and 0.73 for mean indoor CO₂ levels of 1,351 ± 343 ppm and 2,194 ± 623 ppm, respectively. Air analysis indicated that 94 % of dual-source human-material compounds (25 compounds across five categories) accumulated with increasing occupant density. Notably, total emission mass concentrations of dual-origin VOCs significantly exceeded those from materials alone: non-benzenoid hydrocarbons by 10-fold, benzenoid hydrocarbons by 3–4 times, and halogenated compounds/esters by 1.5 times. When the average indoor CO₂ concentration at night was 2,194 ± 623 ppm (with a peak of 2,838 ppm), sleep efficiency declined to 86.3 %, vagal nerve hyperactivity manifested, and post-wake systolic blood pressure increased. To ensure sleep quality and health, average CO₂ concentration of bedroom should remain below 2200 ppm (peak less than 3000 ppm). VOC purification should prioritize non-benzenoid hydrocarbons (less than 65 μg/m³), benzenoid hydrocarbons (less than 225 μg/m³), halogenated compounds (less than 95 μg/m³), and esters (less than 55 μg/m³), with lower concentrations preferred.

有效的睡眠环境空气质量和通风标准的制定取决于人体排放特征的基本数据和对其对睡眠质量影响的清晰认识。本研究量化了三种典型居住者密度下热中性密闭卧室中的CO2和VOC浓度，监测了8名健康成人的睡眠阶段、环境感知和生理反应。结果表明，人均CO2排放量与CO2暴露浓度呈负相关。相对于基线浓度（1063±182 ppm），对于平均室内CO2浓度（1351±343 ppm和2194±623 ppm），这种关系需要分别校正0.79和0.73因子。空气分析表明，随着居住密度的增加，94%的人-物双源化合物（5类25种化合物）积累。值得注意的是，双源挥发性有机化合物的总排放质量浓度明显超过材料本身：非苯类烃的10倍，苯类烃的3-4倍，卤化化合物/酯的1.5倍。当夜间室内平均CO₂浓度为2194±623 ppm（峰值为2838 ppm）时，睡眠效率下降至86.3%，迷走神经表现出过度活跃，醒来后收缩压升高。为了保证睡眠质量和健康，卧室的平均二氧化碳浓度应保持在2200ppm以下（峰值不超过3000ppm）。VOC净化应优先考虑非苯类烃（小于65 μg/m3）、苯类烃（小于225 μg/m3）、卤化化合物（小于95 μg/m3）和酯类（小于55 μg/m3），浓度越低越好。

{"title":"Human metabolic pollutants in bedrooms at typical occupancy density: Composition analysis and effects on sleep quality","authors":"Xiaojing Zhang , Yiting Liu , Mei Wang , Jingchao Xie","doi":"10.1016/j.buildenv.2026.114247","DOIUrl":"10.1016/j.buildenv.2026.114247","url":null,"abstract":"<div><div>The formulation of effective air quality and ventilation standards for sleep environment depends on fundamental data of human emission characteristics and a clear understanding of their impact on sleep quality. This study quantified CO<sub>2</sub> and VOC concentrations in thermally neutral confined bedrooms under three typical occupant densities, monitoring sleep stages, environmental perceptions, and physiological responses in eight healthy adults. Results showed an inverse relationship between per-capita CO<sub>2</sub> emission rate and CO<sub>2</sub> exposure concentration. Relative to the baseline concentration (1,063 ± 182 ppm), this relationship necessitated correction factors of 0.79 and 0.73 for mean indoor CO<sub>2</sub> levels of 1,351 ± 343 ppm and 2,194 ± 623 ppm, respectively. Air analysis indicated that 94 % of dual-source human-material compounds (25 compounds across five categories) accumulated with increasing occupant density. Notably, total emission mass concentrations of dual-origin VOCs significantly exceeded those from materials alone: non-benzenoid hydrocarbons by 10-fold, benzenoid hydrocarbons by 3–4 times, and halogenated compounds/esters by 1.5 times. When the average indoor CO₂ concentration at night was 2,194 ± 623 ppm (with a peak of 2,838 ppm), sleep efficiency declined to 86.3 %, vagal nerve hyperactivity manifested, and post-wake systolic blood pressure increased. To ensure sleep quality and health, average CO<sub>2</sub> concentration of bedroom should remain below 2200 ppm (peak less than 3000 ppm). VOC purification should prioritize non-benzenoid hydrocarbons (less than 65 μg/m<sup>3</sup>), benzenoid hydrocarbons (less than 225 μg/m<sup>3</sup>), halogenated compounds (less than 95 μg/m<sup>3</sup>), and esters (less than 55 μg/m<sup>3</sup>), with lower concentrations preferred.</div></div>","PeriodicalId":9273,"journal":{"name":"Building and Environment","volume":"292 ","pages":"Article 114247"},"PeriodicalIF":7.6,"publicationDate":"2026-01-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146049057","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Effects of clothing types on physiological and perceptual responses in hot and humid underground environments 服装类型对高温潮湿地下环境中生理和知觉反应的影响

IF 7.6 1区工程技术 Q1 CONSTRUCTION & BUILDING TECHNOLOGY

Building and Environment

Pub Date : 2026-01-13 DOI: 10.1016/j.buildenv.2026.114246

Xingming Wang , Yijiang Wang , Wenbo Fan , Chunli Sang , Xiaojuan Lai

The thermal environments encountered by miners in deep underground spaces become more severe as mining depth increases. The miners' work wear (MWW) further amplifies the risk of heat strain. This study aimed to elucidate the extent and mechanisms of physiological and perceptual indicators through a comparative analysis of the thermal responses of subjects wearing MWW versus summer sports clothing (SSC) in a hot and humid environment. Human trials were conducted within a climatic chamber, where subjects' thermophysiological and psychological parameters were measured. Additionally, the rating of perceived exertion (RPE), physiological strain index (PSI), and perceptual strain index (PeSI) were calculated. The results revealed that subjects wearing MWW exhibited heightened thermal physiological and perceptual responses. MWW led to increases in core temperature, mean skin temperature, total sweat loss, and heart rate by 0.35 °C, 0.72 °C, 194.61 g, and 11.82 bpm, compared to SSC conditions, respectively. MWW exacerbated respiratory perception and RPE, and increased the PSI and PeSI. It also diminished the reliability of predicting thermal physiological parameters through thermal perceptions. Furthermore, the rise in metabolic rate amplified the detrimental effects of MWW on salt ion loss, reaching 67.24 mmol/L under MWW versus 12.25 mmol/L under SSC. The impact of MWW on thermal comfort vote was more pronounced than its effect on thermal sensation vote. The findings could provide a foundation for the design of personal protective clothing, the optimization of ventilation strategy, and the management of heat strain in hot and humid underground environments.

随着开采深度的增加，矿工在深部地下空间所遇到的热环境也越来越严峻。矿工的工作服（MWW）进一步加大了热疲劳的风险。本研究旨在通过对比分析穿着MWW和夏季运动服（SSC）的受试者在湿热环境下的热反应，阐明生理和知觉指标的程度和机制。人体试验是在一个气候室中进行的，在那里测量了受试者的热生理和心理参数。此外，计算感知用力评分（RPE）、生理应变指数（PSI）和感知应变指数（PeSI）。结果显示，穿着MWW的受试者表现出更高的热生理和知觉反应。与SSC相比，MWW导致核心温度、平均皮肤温度、总失汗量和心率分别增加0.35°C、0.72°C、194.61 g和11.82 bpm。MWW加重了呼吸感觉和RPE，增加了PSI和PeSI。这也降低了通过热感知预测热生理参数的可靠性。此外，代谢率的升高放大了MWW对盐离子损失的有害影响，MWW下达到67.24 mmol/L，而SSC下为12.25 mmol/L。MWW对热舒适投票的影响大于对热感觉投票的影响。研究结果可为地下湿热环境中个人防护服的设计、通风策略的优化和热应变的管理提供依据。

{"title":"Effects of clothing types on physiological and perceptual responses in hot and humid underground environments","authors":"Xingming Wang , Yijiang Wang , Wenbo Fan , Chunli Sang , Xiaojuan Lai","doi":"10.1016/j.buildenv.2026.114246","DOIUrl":"10.1016/j.buildenv.2026.114246","url":null,"abstract":"<div><div>The thermal environments encountered by miners in deep underground spaces become more severe as mining depth increases. The miners' work wear (MWW) further amplifies the risk of heat strain. This study aimed to elucidate the extent and mechanisms of physiological and perceptual indicators through a comparative analysis of the thermal responses of subjects wearing MWW versus summer sports clothing (SSC) in a hot and humid environment. Human trials were conducted within a climatic chamber, where subjects' thermophysiological and psychological parameters were measured. Additionally, the rating of perceived exertion (RPE), physiological strain index (PSI), and perceptual strain index (PeSI) were calculated. The results revealed that subjects wearing MWW exhibited heightened thermal physiological and perceptual responses. MWW led to increases in core temperature, mean skin temperature, total sweat loss, and heart rate by 0.35 °C, 0.72 °C, 194.61 g, and 11.82 bpm, compared to SSC conditions, respectively. MWW exacerbated respiratory perception and RPE, and increased the PSI and PeSI. It also diminished the reliability of predicting thermal physiological parameters through thermal perceptions. Furthermore, the rise in metabolic rate amplified the detrimental effects of MWW on salt ion loss, reaching 67.24 mmol/L under MWW versus 12.25 mmol/L under SSC. The impact of MWW on thermal comfort vote was more pronounced than its effect on thermal sensation vote. The findings could provide a foundation for the design of personal protective clothing, the optimization of ventilation strategy, and the management of heat strain in hot and humid underground environments.</div></div>","PeriodicalId":9273,"journal":{"name":"Building and Environment","volume":"291 ","pages":"Article 114246"},"PeriodicalIF":7.6,"publicationDate":"2026-01-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146024634","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A quantitative model of indoor illuminance and light color combinations effects on different initial emotional states using physiological measures 室内照度和光色组合对不同初始情绪状态影响的定量模型

IF 7.6 1区工程技术 Q1 CONSTRUCTION & BUILDING TECHNOLOGY

Building and Environment

Pub Date : 2026-01-13 DOI: 10.1016/j.buildenv.2026.114233

Xuejiao Li , Mincheol Whang , Jing Zhang , Hongguo Ren

Quantitative analysis of how light color and illuminance interact to regulate different initial emotions is crucial for improving adaptive intelligent building lighting systems. This study employs event-related potentials (ERPs) from emotion-picture stimuli, brain connectivity metrics, and machine learning methods to investigate the interaction strategies for light color and illuminance in multidimensional emotional regulation across different initial emotion states, spanning physiological to perceptual dimensions. Participants (N = 33) completed the PAD subjective emotion questionnaire and a responsive visual emotion-picture stimuli task under 18 artificial lighting conditions. Results indicate that yellow 180 lx lighting most effectively regulates the valence dimension of negative emotions, while purple 100 lx lighting enhances arousal in positive emotions. Blue lighting at different illuminance levels demonstrates differential effects on attentional focus across emotional states. Illuminance and light color exhibit significant correlations with occipital P300 components and directly influence Fz-Pz brain connectivity metrics. Comparative results of emotion prediction models indicate that using a Multi-Layer Perceptron (MLP) enhances prediction accuracy for both the P300 component and brain connectivity metrics. This study holds significant implications for evaluating emotion and cognition based intelligent lighting systems and researching adaptive lighting systems for emotional regulation. It provides quantitative lighting parameter references for implementing emotional lighting strategies during the initial stages of lighting design.

定量分析光色和照度如何相互作用以调节不同的初始情绪对于改进自适应智能建筑照明系统至关重要。本研究采用事件相关电位（事件相关电位）、脑连通性指标和机器学习方法，研究了不同初始情绪状态下光线颜色和照度在多维情绪调节中的相互作用策略，涵盖生理和感知维度。33名受试者在18种人工照明条件下完成PAD主观情绪问卷和反应性视觉情绪图片刺激任务。结果表明，黄色照明对消极情绪效价维度的调节效果最好，而紫色照明对积极情绪效价维度的调节效果最好。不同照度的蓝色照明对不同情绪状态的注意力集中有不同的影响。照度和光色与枕部P300成分显著相关，并直接影响Fz-Pz脑连接指标。情绪预测模型的比较结果表明，使用多层感知器（MLP）可以提高P300分量和大脑连接指标的预测精度。本研究对于评估基于情绪和认知的智能照明系统以及研究用于情绪调节的自适应照明系统具有重要意义。为在照明设计初期实施情感化照明策略提供定量的照明参数参考。

{"title":"A quantitative model of indoor illuminance and light color combinations effects on different initial emotional states using physiological measures","authors":"Xuejiao Li , Mincheol Whang , Jing Zhang , Hongguo Ren","doi":"10.1016/j.buildenv.2026.114233","DOIUrl":"10.1016/j.buildenv.2026.114233","url":null,"abstract":"<div><div>Quantitative analysis of how light color and illuminance interact to regulate different initial emotions is crucial for improving adaptive intelligent building lighting systems. This study employs event-related potentials (ERPs) from emotion-picture stimuli, brain connectivity metrics, and machine learning methods to investigate the interaction strategies for light color and illuminance in multidimensional emotional regulation across different initial emotion states, spanning physiological to perceptual dimensions. Participants (<em>N</em> = 33) completed the PAD subjective emotion questionnaire and a responsive visual emotion-picture stimuli task under 18 artificial lighting conditions. Results indicate that yellow 180 lx lighting most effectively regulates the valence dimension of negative emotions, while purple 100 lx lighting enhances arousal in positive emotions. Blue lighting at different illuminance levels demonstrates differential effects on attentional focus across emotional states. Illuminance and light color exhibit significant correlations with occipital P300 components and directly influence Fz-Pz brain connectivity metrics. Comparative results of emotion prediction models indicate that using a Multi-Layer Perceptron (MLP) enhances prediction accuracy for both the P300 component and brain connectivity metrics. This study holds significant implications for evaluating emotion and cognition based intelligent lighting systems and researching adaptive lighting systems for emotional regulation. It provides quantitative lighting parameter references for implementing emotional lighting strategies during the initial stages of lighting design.</div></div>","PeriodicalId":9273,"journal":{"name":"Building and Environment","volume":"291 ","pages":"Article 114233"},"PeriodicalIF":7.6,"publicationDate":"2026-01-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146024635","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Addressing overventilation through fan speed reduction in multi-unit residential buildings with pressurized corridor and in-unit ventilation systems 采用加压走廊和单元内通风系统的多单元住宅通过降低风扇转速来解决过度通风问题

IF 7.6 1区工程技术 Q1 CONSTRUCTION & BUILDING TECHNOLOGY

Building and Environment

Pub Date : 2026-01-12 DOI: 10.1016/j.buildenv.2026.114238

Natalie Fylak , Marianne F. Touchie

In-unit ventilation systems (e.g. ERV/HRVs) are becoming increasingly popular in new multi-unit residential buildings (MURBs) in Canada, however centralized ventilation systems are still needed to supply air to the corridors. Corridor ventilation is typically delivered with a pressurized corridor (PC) system, often designed using industry rules of thumb (e.g. 30 cfm /door), resulting in excessively high supply flow rates that could meet the combined corridor and unit ventilation demand. This is an industry holdover from when PC systems were responsible for the whole building ventilation load, and these higher-than-necessary flow rates result in energy waste. A field study was conducted in a MURB located in Toronto, Canada, to determine if PC supply flow rates could be reduced to save energy, while still maintaining corridor pressurization and occupant satisfaction. The study also analysed the impact of occupant behaviors and weatherstripping additions on corridor air pressure changes. During the winter, when stack effect was most severe, the PC system could not maintain positive pressurization on the lowest floors, even at the highest fan speeds. Despite pressurization issues, minimum corridor flow rates required by ASHRAE 62.1 were consistently met and surveys show residents did not perceive changes in air quality as supply flow rates were decreased. Meanwhile, lowering the fan speed to 50% capacity could save 5.5 kWh/m²/year, demonstrating that reducing corridor air supply can be a practical energy saving strategy despite some loss of pressurization on the lowest floors.

单元内通风系统（如ERV/ hrv）在加拿大的新多单元住宅建筑（murb）中越来越受欢迎，但是仍然需要集中通风系统向走廊供气。走廊通风通常采用加压走廊（PC）系统，通常采用行业经验规则（例如30 cfm /门）设计，导致过高的供应流量，可以满足走廊和单元通风的组合需求。这是从PC系统负责整个建筑通风负荷的行业遗留下来的，这些高于必要的流量导致能源浪费。在加拿大多伦多的一个城市城市规划中心进行了一项实地研究，以确定在保持走廊压力和居住者满意度的同时，是否可以降低PC供应流量以节省能源。研究还分析了乘员行为和增加挡风雨条对走廊气压变化的影响。在冬季，当堆栈效应最严重时，PC系统无法在最低楼层保持正压，即使在最高风扇转速下。尽管存在加压问题，但ASHRAE 62.1要求的最低走廊流量始终得到满足，调查显示，随着供应流量的降低，居民并没有感觉到空气质量的变化。同时，将风机转速降低到50%，可以节省5.5 kWh/m²/年，这表明减少走廊送风是一种实用的节能策略，尽管最低楼层的压力会有所损失。

{"title":"Addressing overventilation through fan speed reduction in multi-unit residential buildings with pressurized corridor and in-unit ventilation systems","authors":"Natalie Fylak , Marianne F. Touchie","doi":"10.1016/j.buildenv.2026.114238","DOIUrl":"10.1016/j.buildenv.2026.114238","url":null,"abstract":"<div><div>In-unit ventilation systems (e.g. ERV/HRVs) are becoming increasingly popular in new multi-unit residential buildings (MURBs) in Canada, however centralized ventilation systems are still needed to supply air to the corridors. Corridor ventilation is typically delivered with a pressurized corridor (PC) system, often designed using industry rules of thumb (e.g. 30 cfm /door), resulting in excessively high supply flow rates that could meet the combined corridor and unit ventilation demand. This is an industry holdover from when PC systems were responsible for the whole building ventilation load, and these higher-than-necessary flow rates result in energy waste. A field study was conducted in a MURB located in Toronto, Canada, to determine if PC supply flow rates could be reduced to save energy, while still maintaining corridor pressurization and occupant satisfaction. The study also analysed the impact of occupant behaviors and weatherstripping additions on corridor air pressure changes. During the winter, when stack effect was most severe, the PC system could not maintain positive pressurization on the lowest floors, even at the highest fan speeds. Despite pressurization issues, minimum corridor flow rates required by ASHRAE 62.1 were consistently met and surveys show residents did not perceive changes in air quality as supply flow rates were decreased. Meanwhile, lowering the fan speed to 50% capacity could save 5.5 kWh/m²/year, demonstrating that reducing corridor air supply can be a practical energy saving strategy despite some loss of pressurization on the lowest floors.</div></div>","PeriodicalId":9273,"journal":{"name":"Building and Environment","volume":"291 ","pages":"Article 114238"},"PeriodicalIF":7.6,"publicationDate":"2026-01-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145976332","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Clustering-derived prototypical spectral profiles of glazing for energy, daylighting, and circadian performance 用于能源、采光和昼夜节律性能的玻璃的聚类衍生原型光谱剖面

IF 7.6 1区工程技术 Q1 CONSTRUCTION & BUILDING TECHNOLOGY

Building and Environment

Pub Date : 2026-01-12 DOI: 10.1016/j.buildenv.2026.114239

Neda Ghaeili Ardabili, Julian Wang

Window systems’ spectral profiles influence energy use, daylight availability, and circadian-effective light; however, standard performance metrics rely on single-parameter, integrated values that do not capture the underlying spectral behavior governing these outcomes. This study develops a representative spectral library for architectural glazing through a two-stage clustering framework applied to 6269 spectral power distributions (SPDs) from the International Glazing Database. Using K-means, hierarchical clustering, and Gaussian Mixture Models, and evaluating with the Davies–Bouldin Index, Silhouette Score, WCSS, EVR, and CHI, the dataset is reduced to 25 prototypical SPDs that achieve an average coverage ratio of 70.1% of the spectral patterns in current glazing products. A new circadian-focused design metric, Spatial Circadian Daylight Autonomy (sCDA), is introduced to quantify the spatial and temporal distribution of melanopic daylight. This metric, together with spatial daylight autonomy and ENERGY STAR criteria, is used to assess the representative glazing systems, assembled as insulated glazing units and evaluated across four ENERGY STAR climate zones. Results indicate that the North-Central climate zone is less sensitive to variations in glazing spectral behavior, with 44% of samples meeting all three performance criteria simultaneously. Overall, the study produces a reproducible, data-driven spectral library that supports future research and practical window selection.

窗户系统的光谱分布影响能源使用、日光可用性和昼夜有效光；然而，标准的性能指标依赖于单一参数、综合值，而不能捕获控制这些结果的潜在光谱行为。本研究通过两阶段聚类框架，应用于国际玻璃数据库中的6269个光谱功率分布（spd），开发了一个具有代表性的建筑玻璃光谱库。使用K-means、分层聚类和高斯混合模型，并使用davis - bouldin指数、Silhouette Score、WCSS、EVR和CHI进行评估，将数据集减少到25个原型spd，这些spd的平均覆盖率达到当前玻璃产品中光谱模式的70.1%。引入了一种新的以昼夜节律为中心的设计度量，即空间昼夜节律自主（sCDA），以量化黑视日光的空间和时间分布。该指标与空间日光自主性和能源之星标准一起用于评估代表性的玻璃系统，这些系统被组装为隔热玻璃单元，并在四个能源之星气候区进行评估。结果表明，中北部气候区对玻璃光谱行为的变化不太敏感，44%的样品同时满足所有三个性能标准。总的来说，该研究产生了一个可重复的、数据驱动的光谱库，支持未来的研究和实际的窗口选择。

{"title":"Clustering-derived prototypical spectral profiles of glazing for energy, daylighting, and circadian performance","authors":"Neda Ghaeili Ardabili, Julian Wang","doi":"10.1016/j.buildenv.2026.114239","DOIUrl":"10.1016/j.buildenv.2026.114239","url":null,"abstract":"<div><div>Window systems’ spectral profiles influence energy use, daylight availability, and circadian-effective light; however, standard performance metrics rely on single-parameter, integrated values that do not capture the underlying spectral behavior governing these outcomes. This study develops a representative spectral library for architectural glazing through a two-stage clustering framework applied to 6269 spectral power distributions (SPDs) from the International Glazing Database. Using K-means, hierarchical clustering, and Gaussian Mixture Models, and evaluating with the Davies–Bouldin Index, Silhouette Score, WCSS, EVR, and CHI, the dataset is reduced to 25 prototypical SPDs that achieve an average coverage ratio of 70.1% of the spectral patterns in current glazing products. A new circadian-focused design metric, Spatial Circadian Daylight Autonomy (sCDA), is introduced to quantify the spatial and temporal distribution of melanopic daylight. This metric, together with spatial daylight autonomy and ENERGY STAR criteria, is used to assess the representative glazing systems, assembled as insulated glazing units and evaluated across four ENERGY STAR climate zones. Results indicate that the North-Central climate zone is less sensitive to variations in glazing spectral behavior, with 44% of samples meeting all three performance criteria simultaneously. Overall, the study produces a reproducible, data-driven spectral library that supports future research and practical window selection.</div></div>","PeriodicalId":9273,"journal":{"name":"Building and Environment","volume":"291 ","pages":"Article 114239"},"PeriodicalIF":7.6,"publicationDate":"2026-01-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146024699","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Coupled optimization of double-skin facade design for cold-climate office buildings: Integrating natural ventilation and fire smoke control 冷气候办公建筑双层外立面设计耦合优化：自然通风与火灾防烟一体化

IF 7.6 1区工程技术 Q1 CONSTRUCTION & BUILDING TECHNOLOGY

Building and Environment

Pub Date : 2026-01-12 DOI: 10.1016/j.buildenv.2026.114237

Weiying Chen , Xinyu Zhang , Di Cao , Pengfei Shi , Xin Bi , Jiayin Zhu , Jiantao Zhang , Xinru Jia

Double-skin façade (DSF) are widely applied in office buildings to improve natural ventilation and energy efficiency; however, their unique structure may also accelerate the vertical spread of smoke during a fire. This study develops an integrated optimization framework that simultaneously addresses ventilation performance and fire smoke control. First, baseline datasets—including mean air velocity, ventilation rate, and inner glazing temperature—were generated using Fluent and Pyrosim simulations under various design parameters. Second, an orthogonal analysis was conducted to identify key factors affecting both ventilation enhancement and smoke suppression, focusing on cavity height, cavity width, and vent dimensions. The optimized configurations include: (1) a balanced solution consisting of a two-story cavity, 0.5 m/1.0 m cavity width, 0.8 m exhaust vent, and 1.0 m inlet vent; and (2) a ventilation-prioritized solution consisting of a three-story cavity, 0.5 m/1.0 m cavity width, and 1.0 m exhaust and inlet vents. The proposed framework provides a practical reference for DSF design in similar cold-climate regions, offering improved natural ventilation while ensuring compliance with fire-safety requirements.

双层幕墙（DSF）广泛应用于办公建筑，以改善自然通风和提高能源效率；然而，它们独特的结构也可能在火灾中加速烟雾的垂直蔓延。本研究开发了一个集成的优化框架，同时解决通风性能和火灾烟雾控制。首先，在不同的设计参数下，使用Fluent和Pyrosim模拟生成基线数据集，包括平均空气流速、通风量和内玻璃温度。其次，通过正交分析确定影响通风增强和抑烟的关键因素，重点关注空腔高度、空腔宽度和通风口尺寸。优化后的配置包括：(1)由两层空腔、0.5 m/1.0 m空腔宽度、0.8 m排气孔和1.0 m进气孔组成的平衡溶液；(2)通风优先解决方案，包括三层空腔，空腔宽度为0.5 m/1.0 m，排风口和进风口为1.0 m。提出的框架为类似寒冷气候地区的DSF设计提供了实用参考，在保证防火安全要求的同时提供了更好的自然通风。

{"title":"Coupled optimization of double-skin facade design for cold-climate office buildings: Integrating natural ventilation and fire smoke control","authors":"Weiying Chen , Xinyu Zhang , Di Cao , Pengfei Shi , Xin Bi , Jiayin Zhu , Jiantao Zhang , Xinru Jia","doi":"10.1016/j.buildenv.2026.114237","DOIUrl":"10.1016/j.buildenv.2026.114237","url":null,"abstract":"<div><div>Double-skin façade (DSF) are widely applied in office buildings to improve natural ventilation and energy efficiency; however, their unique structure may also accelerate the vertical spread of smoke during a fire. This study develops an integrated optimization framework that simultaneously addresses ventilation performance and fire smoke control. First, baseline datasets—including mean air velocity, ventilation rate, and inner glazing temperature—were generated using Fluent and Pyrosim simulations under various design parameters. Second, an orthogonal analysis was conducted to identify key factors affecting both ventilation enhancement and smoke suppression, focusing on cavity height, cavity width, and vent dimensions. The optimized configurations include: (1) a balanced solution consisting of a two-story cavity, 0.5 m/1.0 m cavity width, 0.8 m exhaust vent, and 1.0 m inlet vent; and (2) a ventilation-prioritized solution consisting of a three-story cavity, 0.5 m/1.0 m cavity width, and 1.0 m exhaust and inlet vents. The proposed framework provides a practical reference for DSF design in similar cold-climate regions, offering improved natural ventilation while ensuring compliance with fire-safety requirements.</div></div>","PeriodicalId":9273,"journal":{"name":"Building and Environment","volume":"291 ","pages":"Article 114237"},"PeriodicalIF":7.6,"publicationDate":"2026-01-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145975828","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Projecting decarbonization of hospitality buildings in cold climate: Dynamic simulation of hybrid EPS and WWHPs system under shifting energy demands 寒冷气候下酒店建筑脱碳预测：能源需求变化下EPS和WWHPs混合系统的动态模拟

IF 7.6 1区工程技术 Q1 CONSTRUCTION & BUILDING TECHNOLOGY

Building and Environment

Pub Date : 2026-01-11 DOI: 10.1016/j.buildenv.2026.114236

Mohamed G. Ghorab , Antonio Rosato

Climate change is expected to significantly alter building energy demand patterns, particularly in cold regions where heating traditionally dominates. This study addresses the question of how future climatic conditions will impact the energy and environmental performance of advanced geothermal systems in hospitality buildings. To investigate this, a novel hybrid configuration integrating Energy Pile Systems (EPSs) with water-to-water heat pumps (WWHPs) was applied to an 18,000 m² hotel in Ottawa, Canada. Dynamic simulations were conducted using projected climate data for 2050 and 2080 under multiple Representative Concentration Pathways (RCPs). Results indicate a substantial decline in heating demand, with heating degree days decreasing by up to 27.1 %, while cooling requirements rise sharply, with summer cooling degree days increasing by as much as 225.6 %. These shifts influence electrical loads, as Monthly Electrical Peak Intensity (MEPI) decreases during transitional seasons but increases in summer, contributing to an overall reduction in annual electrical Energy Use Intensity (EUI) of up to 4.8 % and a similar decrease in Emission Intensity (EI), contingent on future grid emission factors. The findings underscore the critical role of utility decarbonization in achieving climate targets and highlight the need for adaptive design strategies. This research provides a forward-looking framework for integrating climate resilience and sustainability into hospitality building operations, offering actionable insights for long-term energy planning and environmental impact mitigation.

预计气候变化将显著改变建筑能源需求模式，特别是在传统供暖占主导地位的寒冷地区。本研究解决了未来气候条件将如何影响酒店建筑中先进地热系统的能源和环境性能的问题。为了研究这一点，将能源桩系统（eps）与水对水热泵（WWHPs）集成在一起的新型混合配置应用于加拿大渥太华一家18,000平方米的酒店。利用2050年和2080年的气候预估数据，在多个代表性浓度路径下进行了动态模拟。结果表明，采暖需求大幅下降，采暖度日减少了27.1%，而制冷需求急剧上升，夏季制冷度日增加了225.6%。这些变化会影响电力负荷，因为月度电力峰值强度（MEPI）在过渡季节降低，但在夏季增加，从而导致年度电力能源使用强度（EUI）总体降低高达4.8%，排放强度（EI）也有类似的降低，这取决于未来的电网排放因素。研究结果强调了公用事业脱碳在实现气候目标方面的关键作用，并强调了适应性设计策略的必要性。这项研究为将气候适应能力和可持续性纳入酒店建筑运营提供了一个前瞻性框架，为长期能源规划和减轻环境影响提供了可行的见解。

{"title":"Projecting decarbonization of hospitality buildings in cold climate: Dynamic simulation of hybrid EPS and WWHPs system under shifting energy demands","authors":"Mohamed G. Ghorab , Antonio Rosato","doi":"10.1016/j.buildenv.2026.114236","DOIUrl":"10.1016/j.buildenv.2026.114236","url":null,"abstract":"<div><div>Climate change is expected to significantly alter building energy demand patterns, particularly in cold regions where heating traditionally dominates. This study addresses the question of how future climatic conditions will impact the energy and environmental performance of advanced geothermal systems in hospitality buildings. To investigate this, a novel hybrid configuration integrating Energy Pile Systems (EPSs) with water-to-water heat pumps (WWHPs) was applied to an 18,000 m² hotel in Ottawa, Canada. Dynamic simulations were conducted using projected climate data for 2050 and 2080 under multiple Representative Concentration Pathways (RCPs). Results indicate a substantial decline in heating demand, with heating degree days decreasing by up to 27.1 %, while cooling requirements rise sharply, with summer cooling degree days increasing by as much as 225.6 %. These shifts influence electrical loads, as Monthly Electrical Peak Intensity (MEPI) decreases during transitional seasons but increases in summer, contributing to an overall reduction in annual electrical Energy Use Intensity (EUI) of up to 4.8 % and a similar decrease in Emission Intensity (EI), contingent on future grid emission factors. The findings underscore the critical role of utility decarbonization in achieving climate targets and highlight the need for adaptive design strategies. This research provides a forward-looking framework for integrating climate resilience and sustainability into hospitality building operations, offering actionable insights for long-term energy planning and environmental impact mitigation.</div></div>","PeriodicalId":9273,"journal":{"name":"Building and Environment","volume":"291 ","pages":"Article 114236"},"PeriodicalIF":7.6,"publicationDate":"2026-01-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145975826","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Precooling effectiveness in residential buildings during heatwaves with power outages: A sensitivity and parametric analysis 住宅建筑在热浪断电时预冷效果：灵敏度和参数分析

IF 7.6 1区工程技术 Q1 CONSTRUCTION & BUILDING TECHNOLOGY

Building and Environment

Pub Date : 2026-01-11 DOI: 10.1016/j.buildenv.2026.114235

Xue Liu , Hao Tang , Xuyuan Kang , Pengyuan Shen , Xin Zhou , Da Yan

Climate change is increasing the frequency and intensity of heatwaves, often accompanied by power outages that exacerbate indoor overheating and threaten human health. Precooling is a potential mitigation strategy for indoor overheating risks; however, few studies have systematically explored the impact of building design, cooling capacity, and power-outage events on its performance. This study proposes an optimized precooling thermostat schedule using Bayesian optimization to minimize thermal discomfort and cooling electricity costs. Based on this strategy, Sobol sensitivity and parametric analyses were conducted to explore the effects and interactions of building design parameters, cooling capacity, and power outage characteristics on thermal discomfort and electricity costs. A prototype high-rise residential building in Chengdu, China, was used as a case study. The results showed that thermal discomfort during power outages was primarily influenced by external wall insulation, airtightness and internal thermal mass, whereas cooling electricity cost was mainly affected by airtightness. Furthermore, the contribution of the internal thermal mass to reducing thermal discomfort increased with cooling capacity, whereas the influence of airtightness diminished. This suggests that coordinated optimization between cooling capacity and internal thermal mass is more effective than merely oversizing the cooling system. Regarding power outage impacts, the optimized precooling strategy maintained acceptable indoor comfort for approximately 6 h during morning outages, while only 2 h during afternoon outages. These findings offer practical guidance for policymakers and residents seeking to maximize the benefits of precooling strategies during heatwaves with power outages.

气候变化正在增加热浪的频率和强度，往往伴随着电力中断，加剧室内过热，威胁人类健康。预冷是室内过热风险的潜在缓解策略；然而，很少有研究系统地探讨了建筑设计、冷却能力和停电事件对其性能的影响。本研究提出了一种利用贝叶斯优化的预冷温控器优化计划，以最大限度地减少热不适和冷却电力成本。基于这一策略，进行了Sobol敏感性和参数分析，以探索建筑设计参数、制冷量和停电特性对热不适和电力成本的影响和相互作用。中国成都的一座高层住宅原型建筑被用作案例研究。结果表明：停电时的热不适感主要受外墙保温、气密性和内部热质量的影响，而冷却电费主要受气密性的影响。此外，内部热质量对减少热不适的贡献随着制冷量的增加而增加，而气密性的影响则减小。这表明冷却能力和内部热质量之间的协调优化比仅仅扩大冷却系统更有效。在停电影响方面，优化后的预冷策略在上午停电时可维持约6小时的可接受室内舒适度，而在下午停电时仅维持2小时。这些发现为政策制定者和居民提供了实用的指导，他们希望在停电的热浪中最大化预冷策略的好处。

{"title":"Precooling effectiveness in residential buildings during heatwaves with power outages: A sensitivity and parametric analysis","authors":"Xue Liu , Hao Tang , Xuyuan Kang , Pengyuan Shen , Xin Zhou , Da Yan","doi":"10.1016/j.buildenv.2026.114235","DOIUrl":"10.1016/j.buildenv.2026.114235","url":null,"abstract":"<div><div>Climate change is increasing the frequency and intensity of heatwaves, often accompanied by power outages that exacerbate indoor overheating and threaten human health. Precooling is a potential mitigation strategy for indoor overheating risks; however, few studies have systematically explored the impact of building design, cooling capacity, and power-outage events on its performance. This study proposes an optimized precooling thermostat schedule using Bayesian optimization to minimize thermal discomfort and cooling electricity costs. Based on this strategy, Sobol sensitivity and parametric analyses were conducted to explore the effects and interactions of building design parameters, cooling capacity, and power outage characteristics on thermal discomfort and electricity costs. A prototype high-rise residential building in Chengdu, China, was used as a case study. The results showed that thermal discomfort during power outages was primarily influenced by external wall insulation, airtightness and internal thermal mass, whereas cooling electricity cost was mainly affected by airtightness. Furthermore, the contribution of the internal thermal mass to reducing thermal discomfort increased with cooling capacity, whereas the influence of airtightness diminished. This suggests that coordinated optimization between cooling capacity and internal thermal mass is more effective than merely oversizing the cooling system. Regarding power outage impacts, the optimized precooling strategy maintained acceptable indoor comfort for approximately 6 h during morning outages, while only 2 h during afternoon outages. These findings offer practical guidance for policymakers and residents seeking to maximize the benefits of precooling strategies during heatwaves with power outages.</div></div>","PeriodicalId":9273,"journal":{"name":"Building and Environment","volume":"291 ","pages":"Article 114235"},"PeriodicalIF":7.6,"publicationDate":"2026-01-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145975829","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A neighborhood effect-based method for urban wind environment assessment and regulation 基于邻域效应的城市风环境评价与调控方法

IF 7.6 1区工程技术 Q1 CONSTRUCTION & BUILDING TECHNOLOGY

Building and Environment

Pub Date : 2026-01-10 DOI: 10.1016/j.buildenv.2026.114231

Xiaoqin Liu , Shuliang Zhao , Xiang Huang , Yuqing Ma , Chunyang Zhu , Zongchi Fu

Urban ventilation plays a key role in improving air quality, mitigating the urban heat island effect, and enhancing urban livability. Traditional simulation methods often struggle to balance computational efficiency and cost. To address this, this study proposes a neighborhood-based analytical method (NBAM). The study constructs a dataset of building morphology indicators within neighborhood scales ranging from 1 × 1 to 7 × 7, using the wind speed at the central grid as the target variable. The predictive performance of the MLR model and four machine learning algorithms (XGBoost, RF, LGBM, and SVR) are compared, and the models are interpreted using SHAP. The results show that: (1) Neighborhood effects are crucial for prediction accuracy. Model performance significantly improves as the neighborhood scale expands, with the 7 × 7 neighborhood effectively capturing the influence of surrounding buildings on the wind environment of the target area. (2) The XGBoost algorithm performs excellently, with the model achieving a maximum R² of 0.851 and reducing the mean absolute error to 6.4%. (3) SHAP can be applied in the “Identify-Diagnose-Regulation” analysis process, providing a data-driven decision support tool for fine-tuning urban wind environments. This study not only confirms the necessity of neighborhood effects in urban wind environment research, but also provides a new approach for the precise identification and regulation of urban wind environment issues, as well as for urban ventilation assessment.

城市通风对于改善空气质量、缓解城市热岛效应、提高城市宜居性具有重要作用。传统的仿真方法往往难以平衡计算效率和成本。为了解决这一问题，本研究提出了一种基于社区的分析方法（NBAM）。以中心栅格风速为目标变量，构建了1 × 1 ~ 7 × 7邻域尺度下的建筑形态指标数据集。比较了MLR模型和四种机器学习算法（XGBoost、RF、LGBM和SVR）的预测性能，并使用SHAP对模型进行了解释。结果表明：(1)邻域效应对预测精度至关重要。随着邻域尺度的扩大，模型性能显著提高，7 × 7邻域有效捕捉了周边建筑对目标区域风环境的影响。(2) XGBoost算法性能优异，模型最大R²为0.851，平均绝对误差降至6.4%。(3) SHAP可应用于“识别-诊断-调节”分析过程，为城市风环境的微调提供数据驱动的决策支持工具。本研究不仅确认了邻域效应在城市风环境研究中的必要性，而且为城市风环境问题的精准识别和调控以及城市通风评价提供了新的途径。

{"title":"A neighborhood effect-based method for urban wind environment assessment and regulation","authors":"Xiaoqin Liu , Shuliang Zhao , Xiang Huang , Yuqing Ma , Chunyang Zhu , Zongchi Fu","doi":"10.1016/j.buildenv.2026.114231","DOIUrl":"10.1016/j.buildenv.2026.114231","url":null,"abstract":"<div><div>Urban ventilation plays a key role in improving air quality, mitigating the urban heat island effect, and enhancing urban livability. Traditional simulation methods often struggle to balance computational efficiency and cost. To address this, this study proposes a neighborhood-based analytical method (NBAM). The study constructs a dataset of building morphology indicators within neighborhood scales ranging from 1 × 1 to 7 × 7, using the wind speed at the central grid as the target variable. The predictive performance of the MLR model and four machine learning algorithms (XGBoost, RF, LGBM, and SVR) are compared, and the models are interpreted using SHAP. The results show that: (1) Neighborhood effects are crucial for prediction accuracy. Model performance significantly improves as the neighborhood scale expands, with the 7 × 7 neighborhood effectively capturing the influence of surrounding buildings on the wind environment of the target area. (2) The XGBoost algorithm performs excellently, with the model achieving a maximum R² of 0.851 and reducing the mean absolute error to 6.4%. (3) SHAP can be applied in the “Identify-Diagnose-Regulation” analysis process, providing a data-driven decision support tool for fine-tuning urban wind environments. This study not only confirms the necessity of neighborhood effects in urban wind environment research, but also provides a new approach for the precise identification and regulation of urban wind environment issues, as well as for urban ventilation assessment.</div></div>","PeriodicalId":9273,"journal":{"name":"Building and Environment","volume":"291 ","pages":"Article 114231"},"PeriodicalIF":7.6,"publicationDate":"2026-01-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145976330","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Optimizing Trade-off Between Pollutant Exposure and Energy Consumption in Buildings: Uncertainty Informed Reinforcement Learning and Robustness Analysis of Control Policies 建筑物中污染物暴露与能源消耗之间的优化权衡：不确定性通知强化学习和控制策略的鲁棒性分析

IF 7.6 1区工程技术 Q1 CONSTRUCTION & BUILDING TECHNOLOGY

Building and Environment

Pub Date : 2026-01-10 DOI: 10.1016/j.buildenv.2026.114224

Nishchaya Kumar Mishra , Sameer Patel

Intelligent control systems are crucial for ensuring occupants' comfort and minimizing exposure to indoor pollutants. At the same time, these systems must optimize the building's energy use. Recent research in building science and control has demonstrated the advantages of reinforcement learning (RL) agents over their physics-based and rule-based counterparts in optimizing indoor environment dynamics. While previous studies have shown that RL agents can be transferred across buildings, ensuring their robustness and reliability is essential for such transfer to be effective in real-world applications. Consequently, this study analyzes the robustness of the decision-making ability of a deep Q network (DQN)-based RL agent and estimates the uncertainties in the predicted actions using the Monte Carlo (MC) dropout approach. The performance of twelve action selection policies (MC-DQN1 to MC-DQN12) has been assessed in terms of associated particulate matter (PM) exposure and energy consumption, with the traditional DQN (TradDQN) serving as the benchmark. The analysis reveals substantial variability in average exposure among different MC-DQNs for various emission activities, with variations ranging from -23% to +34% during high-emission activities. Finally, fractional exposures corresponding to different indoor PM levels (≤ 10, 11–20, 21–30, 31–40, and > 40 µg m^-3) have been estimated to identify the specific periods during which these MC-DQNs underperform. Barring a few MC-DQNs that effectively reduced exposure at high PM levels (> 40 µg m^-3), other agents struggled to bring the PM levels to the desired level.

智能控制系统对于确保居住者的舒适和最大限度地减少室内污染物的暴露至关重要。同时，这些系统必须优化建筑的能源使用。最近在建筑科学和控制方面的研究表明，强化学习（RL）智能体在优化室内环境动力学方面优于基于物理和基于规则的智能体。虽然先前的研究表明RL代理可以跨建筑物进行转移，但确保其稳健性和可靠性对于这种转移在实际应用中有效至关重要。因此，本研究分析了基于深度Q网络（DQN）的RL代理的决策能力的鲁棒性，并使用蒙特卡罗（MC） dropout方法估计预测行为中的不确定性。以传统DQN （TradDQN）为基准，对MC-DQN1至MC-DQN12十二个行动选择政策的表现进行了相关颗粒物（PM）暴露和能量消耗的评估。分析显示，不同mc - dqn在各种排放活动下的平均暴露量存在显著差异，在高排放活动期间的变化范围为-23%至+34%。最后，估算了不同室内PM水平（≤10、11-20、21-30、31-40和>； 40µg m-3）对应的暴露分数，以确定这些mc - dqn表现不佳的具体时期。除了少数mc - dqn能有效减少高PM水平（40µg -3）下的暴露外，其他药物难以将PM水平降至所需水平。

{"title":"Optimizing Trade-off Between Pollutant Exposure and Energy Consumption in Buildings: Uncertainty Informed Reinforcement Learning and Robustness Analysis of Control Policies","authors":"Nishchaya Kumar Mishra , Sameer Patel","doi":"10.1016/j.buildenv.2026.114224","DOIUrl":"10.1016/j.buildenv.2026.114224","url":null,"abstract":"<div><div>Intelligent control systems are crucial for ensuring occupants' comfort and minimizing exposure to indoor pollutants. At the same time, these systems must optimize the building's energy use. Recent research in building science and control has demonstrated the advantages of reinforcement learning (RL) agents over their physics-based and rule-based counterparts in optimizing indoor environment dynamics. While previous studies have shown that RL agents can be transferred across buildings, ensuring their robustness and reliability is essential for such transfer to be effective in real-world applications. Consequently, this study analyzes the robustness of the decision-making ability of a deep Q network (DQN)-based RL agent and estimates the uncertainties in the predicted actions using the Monte Carlo (MC) dropout approach. The performance of twelve action selection policies (MC-DQN1 to MC-DQN12) has been assessed in terms of associated particulate matter (PM) exposure and energy consumption, with the traditional DQN (TradDQN) serving as the benchmark. The analysis reveals substantial variability in average exposure among different MC-DQNs for various emission activities, with variations ranging from -23% to +34% during high-emission activities. Finally, fractional exposures corresponding to different indoor PM levels (≤ 10, 11–20, 21–30, 31–40, and > 40 µg m<sup>-3</sup>) have been estimated to identify the specific periods during which these MC-DQNs underperform. Barring a few MC-DQNs that effectively reduced exposure at high PM levels (> 40 µg m<sup>-3</sup>), other agents struggled to bring the PM levels to the desired level.</div></div>","PeriodicalId":9273,"journal":{"name":"Building and Environment","volume":"291 ","pages":"Article 114224"},"PeriodicalIF":7.6,"publicationDate":"2026-01-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145976329","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0