Pub Date : 2025-03-11DOI: 10.1016/j.enbuild.2025.115564
Zehuan Hu , Yuan Gao , Luning Sun , Masayuki Mae , Taiji Imaizumi
Residential buildings account for a significant portion of global energy consumption, making the optimal control of air conditioning and energy systems crucial for improving energy efficiency. However, existing reinforcement learning (RL) methods face challenges, such as the need for carefully designed reward functions in direct RL and the dual training phases required in imitation learning (IL). To address these issues, this study proposes a Generative Adversarial Soft Actor-Critic (GASAC) framework for controlling residential air conditioning and photovoltaic-battery energy storage systems. This framework eliminates the need for predefined reward functions and achieves optimal control through a single training process. An accurate simulation model was developed and validated using real-world data from a residential building in Japan to evaluate the proposed method’s performance. The results show that the proposed method, without requiring a reward function, increased the time the temperature remained within the comfort range by 11.43 % and reduced electricity costs by 14.05 % compared to baseline methods. Additionally, the training time was reduced by approximately two-thirds compared to direct RL methods. These findings demonstrate the effectiveness of GASAC in achieving optimal temperature control and energy savings while addressing the limitations of traditional RL and IL methods.
{"title":"A novel reinforcement learning method based on generative adversarial network for air conditioning and energy system control in residential buildings","authors":"Zehuan Hu , Yuan Gao , Luning Sun , Masayuki Mae , Taiji Imaizumi","doi":"10.1016/j.enbuild.2025.115564","DOIUrl":"10.1016/j.enbuild.2025.115564","url":null,"abstract":"<div><div>Residential buildings account for a significant portion of global energy consumption, making the optimal control of air conditioning and energy systems crucial for improving energy efficiency. However, existing reinforcement learning (RL) methods face challenges, such as the need for carefully designed reward functions in direct RL and the dual training phases required in imitation learning (IL). To address these issues, this study proposes a Generative Adversarial Soft Actor-Critic (GASAC) framework for controlling residential air conditioning and photovoltaic-battery energy storage systems. This framework eliminates the need for predefined reward functions and achieves optimal control through a single training process. An accurate simulation model was developed and validated using real-world data from a residential building in Japan to evaluate the proposed method’s performance. The results show that the proposed method, without requiring a reward function, increased the time the temperature remained within the comfort range by 11.43 % and reduced electricity costs by 14.05 % compared to baseline methods. Additionally, the training time was reduced by approximately two-thirds compared to direct RL methods. These findings demonstrate the effectiveness of GASAC in achieving optimal temperature control and energy savings while addressing the limitations of traditional RL and IL methods.</div></div>","PeriodicalId":11641,"journal":{"name":"Energy and Buildings","volume":"336 ","pages":"Article 115564"},"PeriodicalIF":6.6,"publicationDate":"2025-03-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143644893","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The construction industry is a key stakeholder in mitigating climate change both in China and globally. Understanding the link between spatial network carbon emissions in the construction industry (CECI) and synergy levels in urban agglomerations is crucial for regional emission reduction efforts. This study investigates the Guangdong-Fujian-Zhejiang Coastal Urban Agglomeration (GFZ-CUA) in southeast China, using panel data from 2012 to 2021 to assess CECI. Social network analysis is applied to examine the spatial structure of CECI, supplemented by an analysis of carbon emission coordination discrepancies to empirically reveal both quantitative and relational effects of spatial correlations. Results indicate that the GFZ-CUA exhibits a spatial pattern of “high emissions in the north, low in the south, high along the coast, and low inland,” forming a “core-periphery” structure. Despite broad spatial coverage, relationships among cities frequently shift, revealing fragmented emission reduction efforts between core coastal and peripheral inland cities. Significant disparities are evident in provincial coordination levels. While increasing network density strengthens collaborative emission reduction, the influence of hierarchical structures remains minimal. This study presents a new methodological blueprint for collaborative carbon reduction in urban agglomerations and offers practical policy recommendations and strategies for the low-carbon development of the construction industry.
{"title":"Synergetic carbon emission reduction in the construction industry: A spatial correlation perspective from urban agglomerations in southeast coastal China","authors":"Xiaojuan Li, Gaona Duan, Chengxin Lin, Yun Lin, Jieyi Chen, Congying Fang, Tinghui Zhang","doi":"10.1016/j.enbuild.2025.115552","DOIUrl":"10.1016/j.enbuild.2025.115552","url":null,"abstract":"<div><div>The construction industry is a key stakeholder in mitigating climate change both in China and globally. Understanding the link between spatial network carbon emissions in the construction industry (CECI) and synergy levels in urban agglomerations is crucial for regional emission reduction efforts. This study investigates the Guangdong-Fujian-Zhejiang Coastal Urban Agglomeration (GFZ-CUA) in southeast China, using panel data from 2012 to 2021 to assess CECI. Social network analysis is applied to examine the spatial structure of CECI, supplemented by an analysis of carbon emission coordination discrepancies to empirically reveal both quantitative and relational effects of spatial correlations. Results indicate that the GFZ-CUA exhibits a spatial pattern of “high emissions in the north, low in the south, high along the coast, and low inland,” forming a “core-periphery” structure. Despite broad spatial coverage, relationships among cities frequently shift, revealing fragmented emission reduction efforts between core coastal and peripheral inland cities. Significant disparities are evident in provincial coordination levels. While increasing network density strengthens collaborative emission reduction, the influence of hierarchical structures remains minimal. This study presents a new methodological blueprint for collaborative carbon reduction in urban agglomerations and offers practical policy recommendations and strategies for the low-carbon development of the construction industry.</div></div>","PeriodicalId":11641,"journal":{"name":"Energy and Buildings","volume":"336 ","pages":"Article 115552"},"PeriodicalIF":6.6,"publicationDate":"2025-03-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143644897","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-03-10DOI: 10.1016/j.enbuild.2025.115583
Cun Liu, Chang Xu, Huanxin Chen, Yiqing Wang, Bo Zhu, Linzhi Han
In today’s rapid development of air-conditioning intelligence, blockchain technology has become the key to breaking through the trust barrier to air-conditioning data. Data plays an important role in air-conditioning intelligence. However, the distrust of data has not only hindered the circulation of data, but also limited the further development of air-conditioning intelligence. In order to break through the trust barrier to air-conditioning data, this paper proposes an electronic fault record sharing platform for building air-conditioning based on blockchain technology. The platform utilizes key technologies such as hash algorithm, consensus algorithm, and asymmetric encryption in the blockchain, and chooses the Ethereum network as its deployment platform. The core feature of this platform is its use of blockchain technology to ensure that the data uploaded to the platform is tamper-proof, transparent, and traceable, which achieves completely trustworthy storage within the blockchain and breaks down the trust barrier between individuals and individuals, and between enterprises and enterprises. The results of the study show that smart contracts of the Ethereum platform are excellent in handling electronic fault records, which also provides an entry point for the integration of blockchain with building air-conditioning. In addition, the building air-conditioning electronic fault record sharing platform based on blockchain technology also has a low data storage cost, showing broad prospects for its practical application. Through the application of blockchain technology, the trustworthiness of air-conditioning data is improved, providing a solid foundation for the further development of air-conditioning intelligence.
{"title":"Electronic fault record sharing platform for building air-conditioning based on blockchain technology","authors":"Cun Liu, Chang Xu, Huanxin Chen, Yiqing Wang, Bo Zhu, Linzhi Han","doi":"10.1016/j.enbuild.2025.115583","DOIUrl":"10.1016/j.enbuild.2025.115583","url":null,"abstract":"<div><div>In today’s rapid development of air-conditioning intelligence, blockchain technology has become the key to breaking through the trust barrier to air-conditioning data. Data plays an important role in air-conditioning intelligence. However, the distrust of data has not only hindered the circulation of data, but also limited the further development of air-conditioning intelligence. In order to break through the trust barrier to air-conditioning data, this paper proposes an electronic fault record sharing platform for building air-conditioning based on blockchain technology. The platform utilizes key technologies such as hash algorithm, consensus algorithm, and asymmetric encryption in the blockchain, and chooses the Ethereum network as its deployment platform. The core feature of this platform is its use of blockchain technology to ensure that the data uploaded to the platform is tamper-proof, transparent, and traceable, which achieves completely trustworthy storage within the blockchain and breaks down the trust barrier between individuals and individuals, and between enterprises and enterprises. The results of the study show that smart contracts of the Ethereum platform are excellent in handling electronic fault records, which also provides an entry point for the integration of blockchain with building air-conditioning. In addition, the building air-conditioning electronic fault record sharing platform based on blockchain technology also has a low data storage cost, showing broad prospects for its practical application. Through the application of blockchain technology, the trustworthiness of air-conditioning data is improved, providing a solid foundation for the further development of air-conditioning intelligence.</div></div>","PeriodicalId":11641,"journal":{"name":"Energy and Buildings","volume":"336 ","pages":"Article 115583"},"PeriodicalIF":6.6,"publicationDate":"2025-03-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143637742","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-03-10DOI: 10.1016/j.enbuild.2025.115582
Weilin Li , Rufei Li , Wenhai Sui , Hao Yu , Haichao Gao , Liu Yang
The transition to a renewable energy-centric system necessitates enhanced demand flexibility in buildings. The load characteristics of electrical appliances in residential buildings are influenced by various factors, including seasonal variations and user habits. However, there is a notable lack of comparative research on flexible load control across multiple seasons and day types throughout the year, as well as a scarcity of indexes and their grading applications that can comprehensively describe the multi-dimensional building flexibility characteristics. Therefore, this study focuses on typical residential buildings in Cold region and explores the flexible control effects of various strategies in diverse seasonal and day type scenarios. An innovation “digitized-graded” index system is established, grading index values based on flexibility strength, duration, and maximum reduction, thereby providing a comprehensive understanding of demand flexibility from both quantitative and qualitative perspectives. The findings reveal that different control strategies excel in different seasons and distinct evaluation dimensions. Specifically, the strategy of changing the air conditioning temperature set point performs best in summer, while the intermittent start-stop cycles strategy is more suitable for winter. Furthermore, these strategies dominate in Flexibility Strength Index (FSI) and Flexibility Duration Index (FDI), achieving maximum values of 39.3 % and 14 h, respectively, signifying “extremely strong flexibility” and “extremely long-term flexibility”. Adjusting the lighting intensity is optimal in the transition season, followed by summer, and is least effective in winter. Additionally, shiftable loads using time transfer strategy exhibit a Maximum Reduction Index (MRI) of up to 13.9, indicating “extremely heavy reduction flexibility”.
{"title":"Study on the year-round demand flexibility in residential buildings: Control strategies and quantification methods","authors":"Weilin Li , Rufei Li , Wenhai Sui , Hao Yu , Haichao Gao , Liu Yang","doi":"10.1016/j.enbuild.2025.115582","DOIUrl":"10.1016/j.enbuild.2025.115582","url":null,"abstract":"<div><div>The transition to a renewable energy-centric system necessitates enhanced demand flexibility in buildings. The load characteristics of electrical appliances in residential buildings are influenced by various factors, including seasonal variations and user habits. However, there is a notable lack of comparative research on flexible load control across multiple seasons and day types throughout the year, as well as a scarcity of indexes and their grading applications that can comprehensively describe the multi-dimensional building flexibility characteristics. Therefore, this study focuses on typical residential buildings in Cold region and explores the flexible control effects of various strategies in diverse seasonal and day type scenarios. An innovation “digitized-graded” index system is established, grading index values based on flexibility strength, duration, and maximum reduction, thereby providing a comprehensive understanding of demand flexibility from both quantitative and qualitative perspectives. The findings reveal that different control strategies excel in different seasons and distinct evaluation dimensions. Specifically, the strategy of changing the air conditioning temperature set point performs best in summer, while the intermittent start-stop cycles strategy is more suitable for winter. Furthermore, these strategies dominate in <em>Flexibility Strength Index</em> (<em>FSI</em>) and <em>Flexibility Duration Index</em> (<em>FDI</em>), achieving maximum values of 39.3 % and 14 h, respectively, signifying “extremely strong flexibility” and “extremely long-term flexibility”. Adjusting the lighting intensity is optimal in the transition season, followed by summer, and is least effective in winter. Additionally, shiftable loads using time transfer strategy exhibit a <em>Maximum Reduction Index</em> (<em>MRI</em>) of up to 13.9, indicating “extremely heavy reduction flexibility”.</div></div>","PeriodicalId":11641,"journal":{"name":"Energy and Buildings","volume":"336 ","pages":"Article 115582"},"PeriodicalIF":6.6,"publicationDate":"2025-03-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143620974","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-03-10DOI: 10.1016/j.enbuild.2025.115579
Lina Morkunaite , Darius Pupeikis , Nikolaos Tsalikidis , Marius Ivaskevicius , Fallon Clare Manhanga , Jurgita Cerneckiene , Paulius Spudys , Paraskevas Koukaras , Dimosthenis Ioannidis , Agis Papadopoulos , Paris Fokaides
As global challenges such as climate change and pandemics increasingly disrupt urban systems, the need for efficient and resilient management of energy resources has become critical. The energy used to prepare domestic hot water (DHW) takes a large proportion of residential buildings’ total thermal energy demand. However, it is often overlooked in research due to its stochastic nature and high dependence on user behaviour. This study explores the identification of the crisis and its severity level in the DHW consumption data and the corresponding control actions necessary to mitigate its impact. To identify crisis severity, we utilised the mobility data of retail/recreation activities and transit stations, making the results generalisable for any crisis. In addition, we used power consumption for DHW preparation data from 10 residential apartment buildings located in Kaunas city to develop a machine learning-based hybrid ensembling stacking classifier (ESC) capable of predicting the crisis and its severity level. Finally, we applied principal component analysis (PCA) and k-means clustering to categorise DHW consumption hours throughout the day for each severity level. The results showed that the developed ESC classifier significantly outperforms () the baseline LGBMC classifier (). Combining the classifier with extracted daily consumption patterns and clusters allows the optimisation of control actions on the supply, distribution, and demand side of the DHW system.
{"title":"Efficiency in building energy use: Pattern discovery and crisis identification in hot-water consumption data","authors":"Lina Morkunaite , Darius Pupeikis , Nikolaos Tsalikidis , Marius Ivaskevicius , Fallon Clare Manhanga , Jurgita Cerneckiene , Paulius Spudys , Paraskevas Koukaras , Dimosthenis Ioannidis , Agis Papadopoulos , Paris Fokaides","doi":"10.1016/j.enbuild.2025.115579","DOIUrl":"10.1016/j.enbuild.2025.115579","url":null,"abstract":"<div><div>As global challenges such as climate change and pandemics increasingly disrupt urban systems, the need for efficient and resilient management of energy resources has become critical. The energy used to prepare domestic hot water (DHW) takes a large proportion of residential buildings’ total thermal energy demand. However, it is often overlooked in research due to its stochastic nature and high dependence on user behaviour. This study explores the identification of the crisis and its severity level in the DHW consumption data and the corresponding control actions necessary to mitigate its impact. To identify crisis severity, we utilised the mobility data of retail/recreation activities and transit stations, making the results generalisable for any crisis. In addition, we used power consumption for DHW preparation data from 10 residential apartment buildings located in Kaunas city to develop a machine learning-based hybrid ensembling stacking classifier (ESC) capable of predicting the crisis and its severity level. Finally, we applied principal component analysis (PCA) and k-means clustering to categorise DHW consumption hours throughout the day for each severity level. The results showed that the developed ESC classifier significantly outperforms (<span><math><mrow><msup><mi>R</mi><mn>2</mn></msup><mo>=</mo><mn>0.99</mn></mrow></math></span>) the baseline LGBMC classifier (<span><math><mrow><msup><mi>R</mi><mn>2</mn></msup><mo>=</mo><mn>0.92</mn></mrow></math></span>). Combining the classifier with extracted daily consumption patterns and clusters allows the optimisation of control actions on the supply, distribution, and demand side of the DHW system.</div></div>","PeriodicalId":11641,"journal":{"name":"Energy and Buildings","volume":"336 ","pages":"Article 115579"},"PeriodicalIF":6.6,"publicationDate":"2025-03-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143637354","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Net-zero energy buildings (NZEBs) are increasingly recognized as a key component of bottom-up energy transitions and decarbonization efforts. However, due to the variability in weather conditions driven by climate change, the actual annual performance of an NZEB may deviate from its optimal design targets. To address this challenge, daily energy management can play a crucial role in maintaining the intended energy balance. A major challenge in daily energy management is ensuring compliance with the annual zero-energy constraint. This paper introduces a novel hierarchical approach for NZEB energy management, integrating medium- and short-term energy management models into a two-layer framework. The upper layer allocates monthly energy consumption to individual loads, while the lower layer translates these allocations into daily boundary conditions. Within this framework, daily load scheduling is optimized to minimize energy costs and reduce residents’ discomfort. A data-driven approach is employed to model both controllable and uncontrollable loads based on residents’ lifestyles and climatic indicators. In particular, thermostatically controllable loads are predicted as a linear function of climatic variables, ensuring adaptive and efficient energy management. The proposed hierarchical approach is implemented in a fully electric, single-family residential NZEB located in Gaithersburg, Maryland, US. The results indicate that a traditional energy management algorithm fails to satisfy the annual zero-energy constraint when annual photovoltaic energy generation decreases to 90% of its current value due to climate change. In contrast, the proposed hierarchical approach successfully maintains the net-zero energy condition even when photovoltaic generation falls below 90% of the current level, while keeping residents’ discomfort within acceptable limits.
{"title":"Energy management of net-zero energy buildings: A two-layer hierarchical approach","authors":"Seyyed Reza Ebrahimi , Morteza Rahimiyan , Mohsen Assili , Amin Hajizadeh","doi":"10.1016/j.enbuild.2025.115592","DOIUrl":"10.1016/j.enbuild.2025.115592","url":null,"abstract":"<div><div>Net-zero energy buildings (NZEBs) are increasingly recognized as a key component of bottom-up energy transitions and decarbonization efforts. However, due to the variability in weather conditions driven by climate change, the actual annual performance of an NZEB may deviate from its optimal design targets. To address this challenge, daily energy management can play a crucial role in maintaining the intended energy balance. A major challenge in daily energy management is ensuring compliance with the annual zero-energy constraint. This paper introduces a novel hierarchical approach for NZEB energy management, integrating medium- and short-term energy management models into a two-layer framework. The upper layer allocates monthly energy consumption to individual loads, while the lower layer translates these allocations into daily boundary conditions. Within this framework, daily load scheduling is optimized to minimize energy costs and reduce residents’ discomfort. A data-driven approach is employed to model both controllable and uncontrollable loads based on residents’ lifestyles and climatic indicators. In particular, thermostatically controllable loads are predicted as a linear function of climatic variables, ensuring adaptive and efficient energy management. The proposed hierarchical approach is implemented in a fully electric, single-family residential NZEB located in Gaithersburg, Maryland, US. The results indicate that a traditional energy management algorithm fails to satisfy the annual zero-energy constraint when annual photovoltaic energy generation decreases to 90% of its current value due to climate change. In contrast, the proposed hierarchical approach successfully maintains the net-zero energy condition even when photovoltaic generation falls below 90% of the current level, while keeping residents’ discomfort within acceptable limits.</div></div>","PeriodicalId":11641,"journal":{"name":"Energy and Buildings","volume":"336 ","pages":"Article 115592"},"PeriodicalIF":6.6,"publicationDate":"2025-03-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143637746","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-03-10DOI: 10.1016/j.enbuild.2025.115591
Ehsan Mousavi , Milad Jafari , Walt Vernon , Troy Savage , Charlie Ruschke
This paper examines the historical context and ongoing challenges associated with electrical load sizing in healthcare facilities, focusing on oversizing electrical distribution systems mandated by the National Electrical Code (NEC). While these codes have been designed to ensure safety and reliability, they often result in larger than necessary systems, particularly in healthcare settings where energy intensity is notably high. This study presents new evidence demonstrating that aggressive demand factors for receptacles—specifically, general 180 VA and dedicated 120-volt circuits—can be adjusted to reflect actual load conditions better. We utilized a probabilistic methodology to analyze more than six million electrical load readings collected on 1196 circuits in 14 hospitals in the United States. A total of 196 hospitals were chosen randomly across the United States, and due to the conceived data privacy, measurements were only carried out in facilities that agreed to participate. We present a probability model to analyze and recommend safe design requirements based on actual plug loads in the hospital. The findings suggest a convergence of demand factors as the number of circuits increases, showcasing up to 30% savings in copper and system sizing.
{"title":"Data-driven probabilistic approach to assess electrical plug loads in healthcare facilities","authors":"Ehsan Mousavi , Milad Jafari , Walt Vernon , Troy Savage , Charlie Ruschke","doi":"10.1016/j.enbuild.2025.115591","DOIUrl":"10.1016/j.enbuild.2025.115591","url":null,"abstract":"<div><div>This paper examines the historical context and ongoing challenges associated with electrical load sizing in healthcare facilities, focusing on oversizing electrical distribution systems mandated by the National Electrical Code (NEC). While these codes have been designed to ensure safety and reliability, they often result in larger than necessary systems, particularly in healthcare settings where energy intensity is notably high. This study presents new evidence demonstrating that aggressive demand factors for receptacles—specifically, general 180 VA and dedicated 120-volt circuits—can be adjusted to reflect actual load conditions better. We utilized a probabilistic methodology to analyze more than six million electrical load readings collected on 1196 circuits in 14 hospitals in the United States. A total of 196 hospitals were chosen randomly across the United States, and due to the conceived data privacy, measurements were only carried out in facilities that agreed to participate. We present a probability model to analyze and recommend safe design requirements based on actual plug loads in the hospital. The findings suggest a convergence of demand factors as the number of circuits increases, showcasing up to 30% savings in copper and system sizing.</div></div>","PeriodicalId":11641,"journal":{"name":"Energy and Buildings","volume":"336 ","pages":"Article 115591"},"PeriodicalIF":6.6,"publicationDate":"2025-03-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143628500","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-03-09DOI: 10.1016/j.enbuild.2025.115581
Jiaxin Chen, Nianping Li, Fangning Shi, Wenrui Zheng, Yongga A
Window views have been demonstrated to benefit physical and mental health of indoor personnels in office, so as to enhance their productivity. Whether window views have similar positive effects on indoor sports personnels is also worth exploring. This study conducted a summer field experiment in two gyms with and without window views to evaluate these effects. Twenty participants were required to walk on treadmills set to 4.5 km/h and 6 km/h to simulate moderate and high intensity exercise. Physiological parameters and psychological assessment were recorded during the experiment. The results indicate that when engaging in moderate and high intensity exercises, window views have positive effect on thermal response, with an improvement of 0.31 in thermal comfort ratings, along with a decrease of 0.29 in thermal sensation ratings. Furthermore, window views contribute to participants’ health during exercise, with reduction of 10.42 % and 8.28 % in dizziness, decreases of 4.04 % and 2.52 % in fatigue degrees for moderate and high intensity exercise respectively. In terms of emotions, the positive emotions were influenced more by dynamic mirror vision than natural views in exercise. The results of the study prove the positive effects of window views on indoor personnel’s thermal response and health during moderate and high intensity exercise. This study is beneficial to enrich theoretical research in dynamic thermal comfort and also provides a scientific basis to satisfy the demands of different application requirements of architectural design.
{"title":"Effects of window views on thermal comfort and health during moderate and high intensity exercise: A summer field experiment","authors":"Jiaxin Chen, Nianping Li, Fangning Shi, Wenrui Zheng, Yongga A","doi":"10.1016/j.enbuild.2025.115581","DOIUrl":"10.1016/j.enbuild.2025.115581","url":null,"abstract":"<div><div>Window views have been demonstrated to benefit physical and mental health of indoor personnels in office, so as to enhance their productivity. Whether window views have similar positive effects on indoor sports personnels is also worth exploring. This study conducted a summer field experiment in two gyms with and without window views to evaluate these effects. Twenty participants were required to walk on treadmills set to 4.5 km/h and 6 km/h to simulate moderate and high intensity exercise. Physiological parameters and psychological assessment were recorded during the experiment. The results indicate that when engaging in moderate and high intensity exercises, window views have positive effect on thermal response, with an improvement of 0.31 in thermal comfort ratings, along with a decrease of 0.29 in thermal sensation ratings. Furthermore, window views contribute to participants’ health during exercise, with reduction of 10.42 % and 8.28 % in dizziness, decreases of 4.04 % and 2.52 % in fatigue degrees for moderate and high intensity exercise respectively. In terms of emotions, the positive emotions were influenced more by dynamic mirror vision than natural views in exercise. The results of the study prove the positive effects of window views on indoor personnel’s thermal response and health during moderate and high intensity exercise. This study is beneficial to enrich theoretical research in dynamic thermal comfort and also provides a scientific basis to satisfy the demands of different application requirements of architectural design.</div></div>","PeriodicalId":11641,"journal":{"name":"Energy and Buildings","volume":"336 ","pages":"Article 115581"},"PeriodicalIF":6.6,"publicationDate":"2025-03-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143620982","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-03-09DOI: 10.1016/j.enbuild.2025.115589
Zeyu Wang , Yuelan Hong , Luying Huang , Miaocui Zheng , Hongping Yuan , Ruochen Zeng
Ensemble learning has garnered increasing attention in building energy prediction over the past decade due to its exceptional predictive accuracy. However, there is a lack of systematic reviews that comprehensively analyze its current research status, limitations, and challenges, particularly in the context of large-scale practical applications. To address this gap, this review article systematically evaluates the application of ensemble learning models in building energy prediction. Using the PRISMA method, 82 relevant articles published between 2013 and 2024 in the Web of Science database were analyzed. The findings indicate that heterogeneous ensemble models, which integrate diverse algorithms, and homogeneous ensemble models, which utilize multiple data subsets, both hold significant potential for enhancing prediction accuracy. Specifically, heterogeneous models achieved accuracy improvements ranging from 2.59% to 80.10%, while homogeneous models demonstrated more stable improvements of 3.83% to 33.89%. Nonetheless, the integration of multiple base models increases computational complexity, resulting in higher computation times. Despite this drawback, the improved prediction accuracy, robustness, and generalization capabilities of ensemble models justify the additional computational cost. The review identifies key limitations, including the subjective selection of learning algorithms, the lack of systematic methods for evaluating model diversity, and insufficient exploration of combination strategies. Future research should focus on developing objective criteria for algorithm selection, advancing diversity evaluation techniques, analyzing the effects of combination methods, comparing computational efficiency, and validating the robustness and generalizability of ensemble models. This study offers valuable insights for researchers and practitioners aiming to optimize ensemble learning models in building energy prediction.
{"title":"A comprehensive review and future research directions of ensemble learning models for predicting building energy consumption","authors":"Zeyu Wang , Yuelan Hong , Luying Huang , Miaocui Zheng , Hongping Yuan , Ruochen Zeng","doi":"10.1016/j.enbuild.2025.115589","DOIUrl":"10.1016/j.enbuild.2025.115589","url":null,"abstract":"<div><div>Ensemble learning has garnered increasing attention in building energy prediction over the past decade due to its exceptional predictive accuracy. However, there is a lack of systematic reviews that comprehensively analyze its current research status, limitations, and challenges, particularly in the context of large-scale practical applications. To address this gap, this review article systematically evaluates the application of ensemble learning models in building energy prediction. Using the PRISMA method, 82 relevant articles published between 2013 and 2024 in the Web of Science database were analyzed. The findings indicate that heterogeneous ensemble models, which integrate diverse algorithms, and homogeneous ensemble models, which utilize multiple data subsets, both hold significant potential for enhancing prediction accuracy. Specifically, heterogeneous models achieved accuracy improvements ranging from 2.59% to 80.10%, while homogeneous models demonstrated more stable improvements of 3.83% to 33.89%. Nonetheless, the integration of multiple base models increases computational complexity, resulting in higher computation times. Despite this drawback, the improved prediction accuracy, robustness, and generalization capabilities of ensemble models justify the additional computational cost. The review identifies key limitations, including the subjective selection of learning algorithms, the lack of systematic methods for evaluating model diversity, and insufficient exploration of combination strategies. Future research should focus on developing objective criteria for algorithm selection, advancing diversity evaluation techniques, analyzing the effects of combination methods, comparing computational efficiency, and validating the robustness and generalizability of ensemble models. This study offers valuable insights for researchers and practitioners aiming to optimize ensemble learning models in building energy prediction.</div></div>","PeriodicalId":11641,"journal":{"name":"Energy and Buildings","volume":"335 ","pages":"Article 115589"},"PeriodicalIF":6.6,"publicationDate":"2025-03-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143610156","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-03-09DOI: 10.1016/j.enbuild.2025.115584
Yuewei Li , Bing Dong , Yueming Qiu
Building load profiles are essential in research on building energy management, efficiency, demand response, and grid planning. With the growing adoption of solar photovoltaics (PV) and electric vehicles (EVs), integrated building load profiles are becoming increasingly important for effective management. However, current methods for generating building load profiles focus only on buildings without considering PV and/or EV adoptions. To address this gap, we propose using the conditional generative adversarial network (cGAN), a machine learning technique that generates realistic data conditioned on specific inputs, to create building load profiles that account for PV and EVs. This approach was tested using a smart meter dataset from a major metropolitan area in the southwest United States, containing years of hourly readings from 110 households with PV and EV adoptions. We extracted the key parameters that can describe the generated and real load profiles, and compared their mean and standard deviation to validate the results. KL divergence and FID scores were also used to compare the distributions. The results showed strong alignment between the generated and actual smart meter data across all PV, EV and seasonal conditions. The data under different combinations of PV, EV and weather conditions serve as conditional inputs for the cGAN, allowing it to generate building load profiles that maintain key statistical characteristics for both cooling and heating seasons, and various installation status of PV and EV. Additionally, this method safeguards customer privacy and reduces the effort needed for analyzing occupant behavior and building physics, which are typically required in physics-based energy models.
{"title":"Conditional generative adversarial network (cGAN) for generating building load profiles with photovoltaics and electric vehicles","authors":"Yuewei Li , Bing Dong , Yueming Qiu","doi":"10.1016/j.enbuild.2025.115584","DOIUrl":"10.1016/j.enbuild.2025.115584","url":null,"abstract":"<div><div>Building load profiles are essential in research on building energy management, efficiency, demand response, and grid planning. With the growing adoption of solar photovoltaics (PV) and electric vehicles (EVs), integrated building load profiles are becoming increasingly important for effective management. However, current methods for generating building load profiles focus only on buildings without considering PV and/or EV adoptions. To address this gap, we propose using the conditional generative adversarial network (cGAN), a machine learning technique that generates realistic data conditioned on specific inputs, to create building load profiles that account for PV and EVs. This approach was tested using a smart meter dataset from a major metropolitan area in the southwest United States, containing years of hourly readings from 110 households with PV and EV adoptions. We extracted the key parameters that can describe the generated and real load profiles, and compared their mean and standard deviation to validate the results. KL divergence and FID scores were also used to compare the distributions. The results showed strong alignment between the generated and actual smart meter data across all PV, EV and seasonal conditions. The data under different combinations of PV, EV and weather conditions serve as conditional inputs for the cGAN, allowing it to generate building load profiles that maintain key statistical characteristics for both cooling and heating seasons, and various installation status of PV and EV. Additionally, this method safeguards customer privacy and reduces the effort needed for analyzing occupant behavior and building physics, which are typically required in physics-based energy models.</div></div>","PeriodicalId":11641,"journal":{"name":"Energy and Buildings","volume":"335 ","pages":"Article 115584"},"PeriodicalIF":6.6,"publicationDate":"2025-03-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143591419","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号