Building and Environment最新文献

{"title":"Field practitioners’ perspectives of challenges and practices that hinder effective mould assessment and remediation in Australian residential buildings","authors":"Lisa Coulburn ,&nbsp;Wendy Miller ,&nbsp;Connie Susilawati ,&nbsp;Heike G. Neumeister-Kemp","doi":"10.1016/j.buildenv.2025.114132","DOIUrl":"10.1016/j.buildenv.2025.114132","url":null,"abstract":"<div><div>Like many other countries, the changing climate means that Australia is experiencing changes in rainfall-related weather events, such as increased intensity or prolonged widespread deluges, that are impacting both old and new housing typologies in large geographical expanses in regional and densely populated locations. Water damaged homes provide conditions for mould proliferation, which may impact occupant health. Little is known about how mould in water damaged housing is assessed and remediated by industry professionals. This lack of knowledge inhibits the development and implementation of appropriate protective legislative mechanisms and public health messaging. This study aimed to collate and analyse the current challenges, approaches and practices in mould assessment and remediation, as experienced in the field by mould testing professionals. A combination of purpose, criteria and snowball sampling techniques was used to recruit mould testing professionals who have experience with residential buildings in Australia. Semi-structured interviews were recorded with each participant, and inductive thematic analysis was conducted on all interview transcriptions. Three main themes emerged: hidden moisture, hidden mould and microscopic contamination; an unregulated mould industry; and unrepaired and unremediated buildings. The study points to a need for regulation, standards and training for all professionals in conjunction with centralised information and education for homeowners and the broader community. Research is also required to examine emerging diagnostic tools for detecting hidden mould and moisture. Better mould assessment and standardisation of remediation with quality control may increase the protection of occupants from indoor mould exposure in housing and improve climate resilience in dwellings<em>.</em></div></div>","PeriodicalId":9273,"journal":{"name":"Building and Environment","volume":"290 ","pages":"Article 114132"},"PeriodicalIF":7.6,"publicationDate":"2025-12-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145799548","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}

{"title":"Energy-exergy performance and indoor comfort of a radiant ceiling cooling system with air-to-air sensible heat exchanger in hot and humid climate","authors":"Shikang Wen,&nbsp;Ruoyu You,&nbsp;Qingyan Chen","doi":"10.1016/j.buildenv.2025.114128","DOIUrl":"10.1016/j.buildenv.2025.114128","url":null,"abstract":"<div><div>Building energy use significantly contributes to global carbon emissions. Radiant ceiling panels (RCP) offer an energy-efficient alternative to traditional fan coil units (FCU) by utilizing low-grade thermal sources from low-lift heat pumps. However, the RCP lacks the dehumidification capability, requiring independent humidity control systems typically using dedicated outdoor air systems (DOAS) with high-lift chillers to cool the outdoor air to address latent loads. In hot and humid climates like Hong Kong with dense occupancy, ventilation demands dominate cooling loads, increasing reliance on high-lift chillers and undermining low-lift heat pump benefits. Additionally, overly cooled dehumidification air often causes occupant thermal discomfort. To address these challenges, this study proposed an RCP cooling system integrating an air-to-air sensible heat exchanger (AASHE) with a DOAS to pre-cool incoming outdoor air and reheat overly cooled supply air. Additionally, system performance was optimized by adjusting dehumidification temperature and implementing variable air volume (VAV) control. The proposed system was implemented in a high-humidity, FCU-based office building and evaluated through EnergyPlus and exergy analysis. Results showed that the proposed system reduced indoor discomfort ratios from 54.0 % to under 0.8 %. In addition, the RCP with VAV system reduced energy use by 8.0 % through fan savings with minimal cooling energy impact. The proposed RCP with AASHE system further reduced total energy use by 19.8 % and improved exergy efficiency from 16.1 % to 21.1 % compared to the original FCU system. These findings demonstrate the significant potential of the proposed system for high-density occupancy buildings in hot and humid climates.</div></div>","PeriodicalId":9273,"journal":{"name":"Building and Environment","volume":"290 ","pages":"Article 114128"},"PeriodicalIF":7.6,"publicationDate":"2025-12-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145760753","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}

{"title":"Winter–summer trade-offs and spatial heterogeneity of the urban thermal environment in a cold coastal city: a Local Climate Zone–based in","authors":"Sheng Xu ,&nbsp;Fei Guo ,&nbsp;PeiSheng Zhu ,&nbsp;HongChi Zhang ,&nbsp;WeiSi Li ,&nbsp;ShiYu Miao ,&nbsp;Jun Zhao ,&nbsp;Jing Dong","doi":"10.1016/j.buildenv.2025.114130","DOIUrl":"10.1016/j.buildenv.2025.114130","url":null,"abstract":"<div><div>Global warming and rapid urbanization are intensifying outdoor thermal risks, yet winter–summer trade-offs of the urban thermal environment in cold coastal cities remain poorly quantified at the Local Climate Zone (LCZ) scale. This study aims to clarify seasonal contrasts and spatial heterogeneity of outdoor thermal comfort across typical LCZs in Dalian, a representative cold coastal city, and to compare the performance of the Universal Thermal Climate Index (UTCI) and Physiologically Equivalent Temperature (PET). A machine-learning–corrected LCZ map was generated using k-means++ clustering of multi-source remote-sensing data; eight LCZ types were selected, and multi-site field campaigns in winter and summer provided synchronous meteorological observations from which UTCI and PET were derived. Non-parametric tests and multiple linear regression were employed to detect significant differences and dominant drivers. Compact/high-rise zones (LCZ2, LCZ4) consistently showed the worst thermal conditions; in winter, wind speed was the dominant negative driver of UTCI (regression coefficients −5.69 to −6.62), whereas in summer thermal stress peaked at 12:00–16:00 under strong solar radiation and inter-LCZ contrasts weakened after ∼16:00. Across seasons UTCI achieved higher goodness of fit than PET (R² ≥ 0.97). The results indicate that problematic areas are more extensive and harder to mitigate in winter, suggesting that seasonal adaptation in cold coastal cities should prioritize wintertime interventions and that the LCZ framework is useful for climate-adaptive urban planning and design.</div></div>","PeriodicalId":9273,"journal":{"name":"Building and Environment","volume":"290 ","pages":"Article 114130"},"PeriodicalIF":7.6,"publicationDate":"2025-12-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145760802","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}

{"title":"Dynamic thermal separation performance of segmented recirculating horizontal air curtain during doorway passage in cold storage","authors":"Yanlei Yu ,&nbsp;Yunfei Xia ,&nbsp;Changsheng Cao ,&nbsp;Haiyang Liu ,&nbsp;Zhijian Liu ,&nbsp;Guangyu Cao ,&nbsp;Jun Gao","doi":"10.1016/j.buildenv.2025.114119","DOIUrl":"10.1016/j.buildenv.2025.114119","url":null,"abstract":"<div><div>During sliding door operation and personnel passage, buoyancy-driven mass and heat exchange through the doorway causes substantial refrigeration load in cold storage. This study investigates the thermal exchange through the doorway and the dynamic thermal separation performance of a segmented recirculating horizontal air curtain (SRHAC). To model the transient effects of sliding door and pedestrian movement, computational fluid dynamics (CFD) with dynamic mesh techniques is conducted and validated through small-scale saltwater experiments. Transient passages through doorways with and without SRHAC are simulated, covering sliding-door speeds of 0.3 – 1.5 m/s and pedestrian speeds of 0.6 – 2.4 m/s. Results reveal a strong correlation between transient heat exchange and the passage process. The thermal separation efficiency of SHRAC achieves 75.2 % in the typical situation. During the early opening and late closing stages, the Coanda effect causes the jet adhering to the door panel, enhancing the aerodynamic sealing. However, once the door is more than half open, jet flapping motion occurs, triggering oscillations in transient heat transfer. Rapid door movement further destabilizes the jet, leading to a sharp decrease in efficiency from 72.5 % to 17.3 %. In contrast, disturbances from pedestrian movement dissipate quickly, with the curtain maintaining a stable efficiency of 72.2 % – 74.3 %. Although air curtain performance is sensitive to passage speed, faster door opening and walking reduce the overall passage duration, further decreasing the associated heat exchange by 62.1 % and 74.1 %, respectively. These findings offer quantitative insights and practical strategies for using air curtains to improve the energy efficiency of cold storage facilities.</div></div>","PeriodicalId":9273,"journal":{"name":"Building and Environment","volume":"290 ","pages":"Article 114119"},"PeriodicalIF":7.6,"publicationDate":"2025-12-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145799547","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}

{"title":"An air quality digital twin for real-time outdoor air quality monitoring and prediction","authors":"Yitong Li ,&nbsp;Holly Josephs ,&nbsp;Yaocheng Wu ,&nbsp;Gediminas Mainelis ,&nbsp;Clinton Andrews ,&nbsp;Jie Gong","doi":"10.1016/j.buildenv.2025.114127","DOIUrl":"10.1016/j.buildenv.2025.114127","url":null,"abstract":"<div><div>Respiratory health is closely tied to air quality, making it essential to measure and model air quality. Digital twins provide a powerful approach for air quality monitoring and prediction due to their ability to integrate real-time data, generate air quality predictions, and provide actionable insights. In the City of Elizabeth, New Jersey, poor air quality is driven by heavy industrial activities, dense traffic on major highways, and emissions from the nearby marine terminals and Newark Liberty International Airport. To tackle these issues, this research aimed to develop a digital twin for air quality monitoring and management for the City of Elizabeth. Using LiDAR scans of Housing Authority buildings, Building Information Models (BIM) were created to digitally represent physical structures. A network of outdoor sensors was deployed to capture real-time data on pollutants, including particulate matter (PM₂.₅) and ozone (O₃). Unlike traditional physics-based air quality models that rely on complex mathematical equations and require significant computational resources, this study employed a data-driven approach. By analyzing spatial and temporal patterns in air quality data, this method efficiently generated real-time air quality predictions. Integrating these predictions into digital twins enhances our understanding of air quality dynamics and enables stakeholders to communicate complex information effectively to the public. Furthermore, residents can make informed choices to improve their living conditions, such as determining the best times to open windows, use air filtration systems, or spend time in outdoor environment.</div></div>","PeriodicalId":9273,"journal":{"name":"Building and Environment","volume":"290 ","pages":"Article 114127"},"PeriodicalIF":7.6,"publicationDate":"2025-12-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145838478","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}

{"title":"Integrated effects of personalized heating devices on thermal comfort, skin temperature, and brain networks","authors":"Nan Zhang ,&nbsp;Chao Liu ,&nbsp;Menglin Zhang ,&nbsp;Wenxuan Li ,&nbsp;Lu Liu ,&nbsp;Weijun Gao","doi":"10.1016/j.buildenv.2025.114123","DOIUrl":"10.1016/j.buildenv.2025.114123","url":null,"abstract":"<div><div>This study investigated the combined effects of three types of personalized environmental control systems (PECS)—a heated desk pad (HDP), a heated seat cushion (HSC), and a heated foot-and-leg warmer (HFW)—on thermal comfort, physiological responses, brain functional networks, and cognitive performance under three cold environments (16 °C, 18 °C, and 20 °C). 40 participants were exposed to different PECSs and non-heating (NH) conditions, during which subjective perceptions, physiological signals, and cognitive performance indices were collected. Results showed that, at 16–18 °C, all three PECSs significantly improved both overall and local thermal sensation and comfort, with HSC producing the strongest local effect. HSC also increased skin temperature (+1.88 °C), enhanced parasympathetic nervous activity associated with relaxation, and improved brain functional state. In particular, at 16 °C and 18 °C, compared with NH, HSC reduced brain power by 0.34–1.16 μv², increased clustering coefficients for α and β bands associated with attention (+0.20∼0.30 and +0.19∼0.27, respectively), and improved both global and local efficiency, thereby suggesting enhanced brain integration efficiency. PECSs supported cognitive performance, with HDP showing the most pronounced effect at 18 °C. Under self-regulated conditions, different PECSs exhibited location-specific heating demands. Among them, HSC achieved a 4.0 K compensation with only 22.8 W and showed the lowest comfort–energy performance ratio (5.7 W/K), achieving the optimal balance between thermal comfort, cognitive enhancement, and energy savings. This study provides experimental evidence and design references for personalized thermal comfort interventions and energy-efficient heating strategies in cold environments.</div></div>","PeriodicalId":9273,"journal":{"name":"Building and Environment","volume":"290 ","pages":"Article 114123"},"PeriodicalIF":7.6,"publicationDate":"2025-12-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145838416","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}

{"title":"Indoor temperature and humidity prediction in residential buildings: A review of the white box and black box modeling techniques","authors":"Chima Cyril Hampo ,&nbsp;Leah H. Schinasi ,&nbsp;Simi Hoque","doi":"10.1016/j.buildenv.2025.114125","DOIUrl":"10.1016/j.buildenv.2025.114125","url":null,"abstract":"<div><div>Residential buildings account for a substantial portion of global energy consumption and are critical for ensuring safe and comfortable indoor environments. Accurate prediction of indoor temperature and humidity is essential for thermal comfort, energy efficiency, and occupant health, yet remains methodologically complex. This review synthesizes 91 peer-reviewed studies published between 2019 and 2025 that applied white box (physics-based) and black box (data-driven) modeling approaches to residential buildings. Studies were identified through Google Scholar and Web of Science, screened using defined inclusion and exclusion criteria, and evaluated based on model type, predictive variables, validation method, and performance metrics. White box models, including nodal, zonal, computational fluid dynamics (CFD), and hybrid physics-based frameworks, capture mechanistic heat and moisture transfer processes and have evolved toward more integrated hygrothermal and airflow coupling since 2021. Black box methods, including shallow and deep neural networks, regression models, and hybrid or ensemble architectures, demonstrate high predictive accuracy for short-term and real-time applications, often achieving sub-degree temperature errors and increasingly predicting both temperature and humidity. Comparative findings show that temperature remains the predominant predictive variable, while humidity, though vital for comfort and health, is less frequently modeled. Geographically, studies are concentrated in Europe, East Asia, and North America, with limited representation of tropical and Global South regions. Seasonally, most research has emphasized heating conditions, though recent efforts address cooling, overheating, and mixed-mode ventilation. Evaluation remains fragmented across metrics such as RMSE, MAE, and correlation coefficients, underscoring the need for standardized reporting practices. Overall, physics-based and data-driven approaches are complementary. The former provide interpretability and physical realism, while the latter offer adaptability and scalability. Future research should prioritize coupled heat–moisture modeling, extend analysis to underrepresented climates and housing types, and develop unified validation benchmarks. Advancing along these directions will strengthen predictive reliability and support healthier, more energy-resilient residential environments.</div></div>","PeriodicalId":9273,"journal":{"name":"Building and Environment","volume":"290 ","pages":"Article 114125"},"PeriodicalIF":7.6,"publicationDate":"2025-12-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145799628","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}

{"title":"How geographic barriers mitigate urban PM2.5 pollution: Evidence from 282 cities in China","authors":"Anjian Song ,&nbsp;Zhiyuan Zhang ,&nbsp;Chunguang Hu ,&nbsp;Luyao Wang","doi":"10.1016/j.buildenv.2025.114126","DOIUrl":"10.1016/j.buildenv.2025.114126","url":null,"abstract":"<div><div>Urban air pollution remains one of the most pressing environmental challenges worldwide, yet the role of natural geography in constraining pollutant dispersion has received limited empirical attention. Existing studies have primarily focused on anthropogenic and meteorological determinants of air quality, often overlooking how terrain configuration and external geographic barriers shape the spatial heterogeneity of pollution. This study examines the influence of geographic constraints on PM_2.5 concentrations across 282 Chinese prefecture-level cities, integrating external topographic features with socioeconomic and morphological factors. The indicator <em>nonconvexity</em> is used to capture terrain-induced constraints surrounding built-up areas. Using an optimal-variable specification, we develop both Ordinary Least Squares (OLS) and eXtreme Gradient Boosting (XGBoost) to interpret nonlinear relationships and threshold effects. Results show that <em>nonconvexity</em> exhibits a strong and significant negative association with PM_2.5 concentrations in both models. Furthermore, the strength and direction of socioeconomic effects vary systematically with the level of geographic constraint. In contexts of high geographical constraints, increases in GDP per capita, population, or road density significantly amplify pollution effects. Further interaction effect analysis identifies critical threshold intervals: for instance, a GDP per capita threshold of 6.14 × 10⁴ yuan (95 % confidence interval) and a road density threshold of 6.02 km/km² (95 % confidence interval). By incorporating natural geography into large-scale empirical modeling, this study provides new evidence that physical terrain is a fundamental, structural driver of air-quality disparities across cities. The findings contribute to a deeper theoretical understanding of how geographic form interacts with socioeconomic processes to shape sustainable and resilient urban environments.</div></div>","PeriodicalId":9273,"journal":{"name":"Building and Environment","volume":"289 ","pages":"Article 114126"},"PeriodicalIF":7.6,"publicationDate":"2025-12-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145797555","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}

{"title":"Review and prospect on advanced ventilation strategies for livable and sustainable buildings","authors":"Sheng Zhang ,&nbsp;Zhang Lin","doi":"10.1016/j.buildenv.2025.114122","DOIUrl":"10.1016/j.buildenv.2025.114122","url":null,"abstract":"<div><div>Ventilation systems are widely implemented to provide clean and conditioned air for indoor air quality and thermal comfort, but are also responsible for the substantial energy consumption and carbon emissions of buildings. This study reviews the advanced ventilation strategies to promote the livability and sustainability of buildings. The advanced ventilation strategies are categorized into air distribution methods (<em>i.e.</em>, improvements in existing air distribution method for heating, and newly developed air distribution methods including interactive cascade ventilation, graded ventilation, and adaptive ventilation), air supply methods (<em>i.e.</em>, swirling air supply, vortex-ring air supply, co-flow air supply, targeting air supply, one dimensional dynamic air supply, and two dimensional dynamic air supply), air exit methods, ventilation control methods (<em>i.e.</em>, control for vertically and horizontally non-uniform environments, intermittent demand-controlled ventilation, and occupancy-aided demand-controlled ventilation), and ventilation performance indices (highlighting elevated air velocity for thermal comfort, cooling effect of air movement for energy efficiency, and airborne infection risk control for indoor air quality). Mechanisms underlying the advanced ventilation strategies for high ventilation performance are revealed regarding five-level deepened demands and three extended demands. Research gaps and future work are outlined, targeting occupant demands centered, air utilization efficient, and adaptive advanced ventilation strategies with the aid of artificial intelligence technologies.</div></div>","PeriodicalId":9273,"journal":{"name":"Building and Environment","volume":"289 ","pages":"Article 114122"},"PeriodicalIF":7.6,"publicationDate":"2025-12-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145797226","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}

{"title":"Partially stochastic deep learning with uncertainty quantification for building thermal modeling","authors":"Emma Hannula ,&nbsp;Arttu Häkkinen ,&nbsp;Felipe Uribe ,&nbsp;Antti Solonen ,&nbsp;Jana de Wiljes ,&nbsp;Lassi Roininen","doi":"10.1016/j.buildenv.2025.114118","DOIUrl":"10.1016/j.buildenv.2025.114118","url":null,"abstract":"<div><div>Making the control of building heating systems more energy efficient is crucial for reducing global energy consumption and greenhouse gas emissions. Traditional rule-based control methods use a static, outdoor temperature-dependent heating curve to regulate heat input. This open-loop approach fails to account for both the current state of the system (indoor temperature) and free heat gains, such as solar radiation, often resulting in poor thermal comfort and overheating. Model Predictive Control (MPC) addresses these drawbacks by using predictive modeling to optimize heating based on a building’s learned thermal behavior, current system state, and weather forecasts. However, current industrial MPC solutions often employ simplified physics-inspired indoor temperature models, sacrificing accuracy for robustness and interpretability. While purely data-driven models offer superior predictive performance and therefore more accurate control, they face challenges such as a lack of transparency.</div><div>To bridge this gap, we propose a partially stochastic deep learning (DL) architecture, dubbed LSTM+BNN, for building-specific indoor temperature modeling. Unlike most studies that evaluate model performance through simulations or limited test buildings, our experiments across a comprehensive dataset of 100 real-world buildings, under various weather conditions, demonstrate that LSTM+BNN outperforms an industry-proven reference model, reducing the average prediction error measured as RMSE by more than 40 % for the 48-hour prediction horizon of interest. Unlike deterministic DL approaches, LSTM+BNN offers a critical advantage by enabling pre-assessment of model competency for control optimization through uncertainty quantification. Thus, the proposed model shows significant potential to improve thermal comfort and energy efficiency achieved with heating MPC solutions.</div></div>","PeriodicalId":9273,"journal":{"name":"Building and Environment","volume":"289 ","pages":"Article 114118"},"PeriodicalIF":7.6,"publicationDate":"2025-12-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145797554","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}