Energy and Buildings最新文献_第4页

Decision-making for renovating the Mediterranean social housing: A practical approach through an interactive open access tool

IF 6.6 2区工程技术

Energy and Buildings Pub Date : 2025-03-21 DOI: 10.1016/j.enbuild.2025.115629

C.M. Calama-González , R. Escandón , R. Suárez , F. Ascione

{"title":"Decision-making for renovating the Mediterranean social housing: A practical approach through an interactive open access tool","authors":"C.M. Calama-González , R. Escandón , R. Suárez , F. Ascione","doi":"10.1016/j.enbuild.2025.115629","DOIUrl":"10.1016/j.enbuild.2025.115629","url":null,"abstract":"<div><div>To achieve 2050 Climate Neutrality, building stock requires a multidimensional renovation process. This is particularly urgent in most vulnerable households, with higher exposure to climate change, where this procedure should focus on cost-controlled passive measures. Given the complexity of identifying optimal strategies, it is imperative to improve the retrofitting process of the social housing stock to enhance its energy performance guaranteeing health and comfort. For this, an interactive tool was developed focused on the case of southern Spain. Able to provide optimized combinations of energy retrofit strategies, using NSGA-II genetic algorithms and setting two optimization objectives: minimizing thermal discomfort and economic costs. The freely accessible tool was designed with practical and didactic approach to facilitate decision-making. The results obtained suggest the feasibility of implementing phase actions instead of a single large-scale intervention and show the tool’s ability to quantify the percentage of thermal comfort improvement achieves at each phase.</div></div>","PeriodicalId":11641,"journal":{"name":"Energy and Buildings","volume":"336 ","pages":"Article 115629"},"PeriodicalIF":6.6,"publicationDate":"2025-03-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143687587","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}

引用次数: 0

An open control sequence specification to scale building demand flexibility via analytics software

IF 6.6 2区工程技术

Energy and Buildings Pub Date : 2025-03-20 DOI: 10.1016/j.enbuild.2025.115616

Jessica Granderson, Armando Casillas, Marco Pritoni, Weiping Huang

{"title":"An open control sequence specification to scale building demand flexibility via analytics software","authors":"Jessica Granderson, Armando Casillas, Marco Pritoni, Weiping Huang","doi":"10.1016/j.enbuild.2025.115616","DOIUrl":"10.1016/j.enbuild.2025.115616","url":null,"abstract":"<div><div>For over two decades, researchers and practitioners have showcased the ability of large commercial buildings to provide grid services by shedding or shifting load. Various utility demand response (DR) and virtual power plant (VPP) programs throughout the United States are presently utilizing these demand-side resources. However, growth of these programs have been limited, in part due to the high cost necessary to integrate the DR control strategies into the building automation system (BAS). Implementing these strategies involves adjusting control sequences, necessitating dozens of hours of customized programming per building, limiting their adoption to large organizations and progressive owners.</div><div>Recent efforts by researchers and industry have demonstrated the capability of energy management and information systems (EMIS), originally designed for fault detection and diagnostics, to interface with existing BAS and perform supervisory control to optimize building operations. While these approaches are quickly being adopted by industry, demand flexibility (DF) control strategies remain limited in product offerings. One of the challenges is the lack of documented best-practice DF sequences, despite the rich literature on field implementations.</div><div>This paper develops a new open-specification for a zone-based temperature adjustment shed strategy for commercial building HVAC systems, describing the specification’s implementation in two EMIS tools in both experimental and field settings. Both implementations successfully reduced electric load by at least 40% on average during the called event, while maintaining temperature limits. This study’s detailed process from specification to deployment shows the potential for scalability as well as highlights challenges related to integration with heterogeneous BAS products.</div></div>","PeriodicalId":11641,"journal":{"name":"Energy and Buildings","volume":"337 ","pages":"Article 115616"},"PeriodicalIF":6.6,"publicationDate":"2025-03-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143739824","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}

引用次数: 0

RC parameter identification and load aggregation analysis of air-conditioning systems: A multi-strategy improved black-winged kite algorithm

IF 6.6 2区工程技术

Energy and Buildings Pub Date : 2025-03-20 DOI: 10.1016/j.enbuild.2025.115641

Mengran Zhou , Chunchen Shi , Feng Hu , Ziwei Zhu , Kun Wang , Xiangnan Sun , Yu Zhang , Mengcheng Zhou , Lehan Zhang , Yuewen Zhang

{"title":"RC parameter identification and load aggregation analysis of air-conditioning systems: A multi-strategy improved black-winged kite algorithm","authors":"Mengran Zhou , Chunchen Shi , Feng Hu , Ziwei Zhu , Kun Wang , Xiangnan Sun , Yu Zhang , Mengcheng Zhou , Lehan Zhang , Yuewen Zhang","doi":"10.1016/j.enbuild.2025.115641","DOIUrl":"10.1016/j.enbuild.2025.115641","url":null,"abstract":"<div><div>As the proportion of air-conditioning loads in power systems continues to increase, their potential as demand response resources is becoming increasingly significant. However, the heterogeneity and dynamic nonlinear characteristics of air-conditioning loads, driven by variations in building environments and user behaviors, often result in insufficient accuracy in traditional parameter identification and aggregation modeling. To address this issue, this study proposes a multi-strategy modified Black-winged Kite Algorithm (MBKA) combined with a first-order Equivalent Thermal Parameter (ETP) model and measured data to identify air-conditioning R and C parameters accurately. Furthermore, the effects of setpoint temperature and initial indoor temperature diversity on aggregation characteristics are analyzed. The results demonstrate that MBKA significantly enhances model identification accuracy, achieving a mean square error (MSE) as low as 0.005860. When considering both setpoint and initial indoor temperature diversity, the volatility of aggregated power is significantly reduced, with the peak-to-average ratio, standard deviation, and coefficient of variation decreasing by 15.83 %, 78.21 %, and 74.43 %, respectively. When only initial indoor temperature diversity is considered, these metrics decrease by 11.18 %, 66.73 %, and 64.95 %, respectively. Additionally, a setpoint temperature-adjustable capacity fitting model is established, exhibiting a high fitting accuracy with an R<sup>2</sup> value of 0.999. This study provides theoretical and technical support for integrating air-conditioning loads into the flexible scheduling of modern power systems through algorithmic improvements and comprehensive aggregation characterization.</div></div>","PeriodicalId":11641,"journal":{"name":"Energy and Buildings","volume":"337 ","pages":"Article 115641"},"PeriodicalIF":6.6,"publicationDate":"2025-03-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143704796","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}

引用次数: 0

Evolutionary game analysis on governments and developers’ behavioral strategies: Impact of dynamic incentives for green building

IF 6.6 2区工程技术

Energy and Buildings Pub Date : 2025-03-20 DOI: 10.1016/j.enbuild.2025.115631

Xinhai Lu , Chuan Yang , Wangrong Ma , Hao Yang

{"title":"Evolutionary game analysis on governments and developers’ behavioral strategies: Impact of dynamic incentives for green building","authors":"Xinhai Lu , Chuan Yang , Wangrong Ma , Hao Yang","doi":"10.1016/j.enbuild.2025.115631","DOIUrl":"10.1016/j.enbuild.2025.115631","url":null,"abstract":"<div><div>Many countries have implemented economic plan and subsidy policies to encourage green building development. This paper constructs an evolutionary game model between governments and developers to examine the impact of government-led incentives on the decision-making process of developers within the dynamic trajectory of green building industry. Considering the purchasing preference of homebuyers, this study aims at to evaluate the impact of incentive policies on the transition to green building, examining three distinct scenarios: the static incentives, the dynamic taxation and the dynamic subsidies. The evolutionary stable strategy between governments and developers is derived. Then, a numerical simulation is employed to demonstrate the dynamic evolution process. The findings indicate that there is no evolutionary stable strategy under static scenario. When the government actively supervises the transition of green buildings and adopts dynamic incentive measures, the evolutionary game demonstrates stability. Furthermore, the simulation result shows that a policy of dynamic subsidies is more effective for the transition to green buildings than alternative incentive strategies. These findings offer a foundation for policymakers to facilitate the transition to green building.</div></div>","PeriodicalId":11641,"journal":{"name":"Energy and Buildings","volume":"336 ","pages":"Article 115631"},"PeriodicalIF":6.6,"publicationDate":"2025-03-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143687522","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}

引用次数: 0

Predicting summer indoor temperatures in Nordic apartments considering heatwaves forecasts

IF 6.6 2区工程技术

Energy and Buildings Pub Date : 2025-03-19 DOI: 10.1016/j.enbuild.2025.115630

Azin Velashjerdi Farahani , Matti Leinonen , Laura Ruotsalainen , Juha Jokisalo , Risto Kosonen

{"title":"Predicting summer indoor temperatures in Nordic apartments considering heatwaves forecasts","authors":"Azin Velashjerdi Farahani , Matti Leinonen , Laura Ruotsalainen , Juha Jokisalo , Risto Kosonen","doi":"10.1016/j.enbuild.2025.115630","DOIUrl":"10.1016/j.enbuild.2025.115630","url":null,"abstract":"<div><div>With global ambient air temperature rise,more intense, frequent, and much longer heatwaves are expected. This is associated with high indoor temperatures and overheating risks in apartments in the Nordic region. This study investigates the prediction of indoor temperatures in Nordic apartment buildings during summer heatwaves using outdoor weather forecasts to forecast indoor overheating risks. A comprehensive dataset of hourly indoor temperatures from over 20,000 apartments in the Helsinki region between 2018 and 2021 was used. The outdoor hourly parameters, including air temperature, humidity, and solar irradiance, were integrated with limited apartment characteristics such as size, number of rooms, and age. Three machine learning models—Long Short-Term Memory (LSTM), XGBoost, and Multivariate Linear Regression (MLR)—were employed to predict indoor temperatures for the next 24, 48, and 72 h. The performance of the models was evaluated using mtrics e.g., Mean Absolute Error (MAE) and Mean Squared Error (MSE. The XGBoost model achieved the highest accuracy with an MAE of 0.23 °C and MSE of 0.12 for the 24-h prediction. While LSTM showed superior performance under high-temperature conditions, compared to XGBoost, it could not capture hourly changes in indoor temperature and follow its pattern. The results provide insights into the challenges of predicting indoor temperatures in residential buildings during extreme heat events and highlight the importance of including outdoor weather forecasts for improved model accuracy. Its findings offer a novel approach to maintaining comfortable indoor thermal conditions and early forecasting ofpossible overheating in Nordic apartments during hot summers.</div></div>","PeriodicalId":11641,"journal":{"name":"Energy and Buildings","volume":"336 ","pages":"Article 115630"},"PeriodicalIF":6.6,"publicationDate":"2025-03-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143687526","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}

引用次数: 0

A data-driven building thermal zoning algorithm for digital twin-enabled advanced control

IF 6.6 2区工程技术

Energy and Buildings Pub Date : 2025-03-19 DOI: 10.1016/j.enbuild.2025.115633

Lina Morkunaite , Adil Rasheed , Darius Pupeikis , Vangelis Angelakis , Tobias Davidsson

{"title":"A data-driven building thermal zoning algorithm for digital twin-enabled advanced control","authors":"Lina Morkunaite , Adil Rasheed , Darius Pupeikis , Vangelis Angelakis , Tobias Davidsson","doi":"10.1016/j.enbuild.2025.115633","DOIUrl":"10.1016/j.enbuild.2025.115633","url":null,"abstract":"<div><div>Effective control of indoor environments is crucial for maintaining occupant comfort and optimizing energy use. However, current building control strategies often fail to achieve these goals, as they rely on static or rule-based approaches that normally do not account for dynamic conditions. While advanced control strategies offer a more adaptive solution, their implementation is challenging due to the need for accurate thermal models, which are resource-intensive to develop. Defining building thermal zones can help to strike a balance between model accuracy and the cost of their development and implementation. However, data-driven approaches for identifying thermal zones remain scarce. This study addresses these gaps by proposing a reusable data-driven thermal zoning algorithm that employs Principal Component Analysis (PCA) and k-means clustering to define building thermal zones. This method allows for the inclusion of numerous parameters, thus increasing the applicability and consistency of the zoning process. Additionally, we propose an algorithm for zones validation, supported by qualitative criteria from literature and standards. The approach is tested in a large educational building, using time-series data from 168 rooms with a total of 262 CO2 and temperature sensors. Results show that the proposed zoning algorithm achieves over 91 % consistency score, depending on the number of selected principal components, clusters, and input parameters available. The derived thermal zones are further validated based on the synthesised qualitative criteria. Finally, the results are visualized in a DT environment, where users can explore color-coded thermal zones alongside real-time sensor data, 3D building geometry, and semantic information.</div></div>","PeriodicalId":11641,"journal":{"name":"Energy and Buildings","volume":"336 ","pages":"Article 115633"},"PeriodicalIF":6.6,"publicationDate":"2025-03-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143687525","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}

引用次数: 0

Advancing energy renovations through digitalisation: A critical review of EU policies and instruments

IF 6.6 2区工程技术

Energy and Buildings Pub Date : 2025-03-18 DOI: 10.1016/j.enbuild.2025.115627

Sun-Ah Hwang, Sultan Çetin, Henk Visscher, Ad Straub

{"title":"Advancing energy renovations through digitalisation: A critical review of EU policies and instruments","authors":"Sun-Ah Hwang, Sultan Çetin, Henk Visscher, Ad Straub","doi":"10.1016/j.enbuild.2025.115627","DOIUrl":"10.1016/j.enbuild.2025.115627","url":null,"abstract":"<div><div>This paper provides a holistic overview of the evolution of policies towards digitalising energy renovation processes in the European Union (EU). Since the European Green Deal initiative in 2019, EU policies have been increasingly addressing the digitalisation of the building industry to enable evidence-based decisions when tackling environmental challenges. To better understand the development(s) since, this paper integrates a structured policy analysis approach and critically reviews 31 EU policy documents on digitalisation and/or energy renovation. The analysis identified a growing number of policy instruments aimed at supporting a robust use of data, to, among others, improve decision-making and information sharing throughout the energy renovation process. These include Energy Performance Certificate, Building Renovation Passport, Smart Readiness Indicators, Level(s), Digital Building Logbook, Digital Product Passport, Digital Twin, Building Information Modelling, and Digital Permitting. While each of these nine instruments can independently facilitate decision-making on sustainable and/or smart renovations, they also project a significant degree of complementarity between each other. To that, this paper presents the Digital Energy Renovation Framework, which comprehensively synthesises the (inter)relationships between the proposed policy instruments with respect to facilitating energy renovation processes. A key finding is that the integration between the Building Renovation Passport and a data-rich Digital Building Logbook is fundamental to maximise the impact on decision-making throughout the renovation process. To achieve this, ensuring coherence and interoperability of data throughout the renovation value chain is crucial, with the standardisation of data formats and protocols being essential for effective data gathering and processing across these instruments.</div></div>","PeriodicalId":11641,"journal":{"name":"Energy and Buildings","volume":"336 ","pages":"Article 115627"},"PeriodicalIF":6.6,"publicationDate":"2025-03-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143687527","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}

引用次数: 0

Ensuring real-time data integrity in smart building applications: A systematic end-to-end comprehensive pipeline evaluated in numerous real-life cases

IF 6.6 2区工程技术

Energy and Buildings Pub Date : 2025-03-18 DOI: 10.1016/j.enbuild.2025.115586

Aliki Stefanopoulou , Iakovos Michailidis , Georgios Karatzinis , Georgios Lepidas , Yiannis S. Boutalis , Elias B. Kosmatopoulos

{"title":"Ensuring real-time data integrity in smart building applications: A systematic end-to-end comprehensive pipeline evaluated in numerous real-life cases","authors":"Aliki Stefanopoulou , Iakovos Michailidis , Georgios Karatzinis , Georgios Lepidas , Yiannis S. Boutalis , Elias B. Kosmatopoulos","doi":"10.1016/j.enbuild.2025.115586","DOIUrl":"10.1016/j.enbuild.2025.115586","url":null,"abstract":"<div><div>In this study, we propose a comprehensive, end-to-end data healing pipeline, developed and tested in real buildings facing diverse severity problems. This pipeline is designed to ensure the accuracy and reliability of smart building data in a fast and responsive manner, using computationally lightweight statistical approaches for outlier detection and LightGBM for data imputation. Both methods are optimized for low computational cost, making them ideal for real-world scenarios requiring immediate feedback and capable of handling very large datasets efficiently. Our system is designed to operate automatically, capable of applying real-time data processing and periodic model updates without manual intervention. It was evaluated using Key Performance Indicators over nine weeks across five smart buildings in the EU, revealing discrepancies in performance across different time periods and buildings. These findings highlight the need for tailored data healing strategies for varying dataset sizes, ultimately enhancing data quality for more reliable analyses and informed decision-making. The implementation of this pipeline contributes to more accurate energy usage, improved occupant comfort, and more efficient building operations, supporting the broader goals of sustainability and energy efficiency in smart buildings.</div></div>","PeriodicalId":11641,"journal":{"name":"Energy and Buildings","volume":"336 ","pages":"Article 115586"},"PeriodicalIF":6.6,"publicationDate":"2025-03-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143687524","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}

引用次数: 0

Multi-energy systems fast optimization: A new formulation in linear programming for temperatures and magnitudes of thermal power flows in heating systems

IF 6.6 2区工程技术

Energy and Buildings Pub Date : 2025-03-18 DOI: 10.1016/j.enbuild.2025.115618

Jonathan Hachez , Arnaud Latiers , Benjamin Berger , Svend Bram

{"title":"Multi-energy systems fast optimization: A new formulation in linear programming for temperatures and magnitudes of thermal power flows in heating systems","authors":"Jonathan Hachez , Arnaud Latiers , Benjamin Berger , Svend Bram","doi":"10.1016/j.enbuild.2025.115618","DOIUrl":"10.1016/j.enbuild.2025.115618","url":null,"abstract":"<div><div>Geopolitical events and environmental pressures can force regions to speed up their energy transition, as seen in the European Union (EU) shift towards sustainable smart energy systems in response to its current socio-economic and geopolitical situation and to climate change. Among the assets that are envisaged are District Heating Network (DHN), Heat Pump (HP), Thermal Energy Storage (TES), and Photovoltaic Panels (PV).To ensure optimal equipment sizing, the widely adopted method is Mixed Integer Linear Programming (MILP) with a fixed supply temperature. Unfortunately, using MILP to optimize the design and operation considering variable supply temperature is not done in the literature because it is time-consuming. However, this article’s Linear Programming (LP) formulation uses temperature levels to optimize the supply temperature and reaches optimality within 20 min. It improves the Coefficient of Performance (COP) of HP and increases the energetic density of TES. A specific case study examines a residential building with PV connected to a <span><math><msup><mn>5</mn><mtext>th</mtext></msup></math></span> generation DHN, showing that electric self-production can reach 58 %, Seasonal Coefficient of Performance (SCOP) stands at 4.1, and electric self-consumption rate reaches 81 %. This formulation is more complete for optimizing low-temperature DHN, as it accounts for the sensitivity of equipment performance to network temperatures, considering the availability of renewable energy sources.</div></div>","PeriodicalId":11641,"journal":{"name":"Energy and Buildings","volume":"336 ","pages":"Article 115618"},"PeriodicalIF":6.6,"publicationDate":"2025-03-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143687586","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}

引用次数: 0

Transfer learning on transformers for building energy consumption forecasting—A comparative study

IF 6.6 2区工程技术

Energy and Buildings Pub Date : 2025-03-18 DOI: 10.1016/j.enbuild.2025.115632

Robert Spencer , Surangika Ranathunga , Mikael Boulic , Andries (Hennie) van Heerden , Teo Susnjak

{"title":"Transfer learning on transformers for building energy consumption forecasting—A comparative study","authors":"Robert Spencer , Surangika Ranathunga , Mikael Boulic , Andries (Hennie) van Heerden , Teo Susnjak","doi":"10.1016/j.enbuild.2025.115632","DOIUrl":"10.1016/j.enbuild.2025.115632","url":null,"abstract":"<div><div>Energy consumption in buildings is steadily increasing, leading to higher carbon emissions. Predicting energy consumption is a key factor in addressing climate change. There has been a significant shift from traditional statistical models to advanced deep learning (DL) techniques for predicting energy use in buildings. However, data scarcity in newly constructed or poorly instrumented buildings limits the effectiveness of standard DL approaches. In this study, we investigate the application of six data-centric Transfer Learning (TL) strategies on three Transformer architectures—vanilla Transformer, Informer, and PatchTST—to enhance building energy consumption forecasting. Transformers, a relatively new DL framework, have demonstrated significant promise in various domains; yet, prior TL research has often focused on either a single data-centric strategy or older models such as Recurrent Neural Networks. Using 16 diverse datasets from the Building Data Genome Project 2, we conduct an extensive empirical analysis under varying feature spaces (e.g., recorded ambient weather) and building characteristics (e.g., dataset volume). Our experiments show that combining multiple source datasets under a zero-shot setup reduces the Mean Absolute Error (MAE) of the vanilla Transformer model by an average of 15.9 % for 24 h forecasts, compared to single-source baselines. Further fine-tuning these multi-source models with target-domain data yields an additional 3–5 % improvement. Notably, PatchTST outperforms the vanilla Transformer and Informer models. Overall, our results underscore the potential of combining Transformer architectures with TL techniques to enhance building energy consumption forecasting accuracy. However, careful selection of the TL strategy and attention to feature space compatibility are needed to maximize forecasting gains.</div></div>","PeriodicalId":11641,"journal":{"name":"Energy and Buildings","volume":"336 ","pages":"Article 115632"},"PeriodicalIF":6.6,"publicationDate":"2025-03-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143687523","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}

引用次数: 0