Machine learning for estimation of building energy consumption and performance: a review

In recent years, Artificial Intelligence (AI) in general and Machine Learning (ML) techniques in specific terms have been proposed for forecasting of building energy consumption and performance. This chapter provides a substantial review on the four main ML approaches including artificial neural network, support vector machine, Gaussian-based regressions and clustering, which have commonly been applied in forecasting and improving building energy performance.

Buildings account for nearly 40% of the global process and industrial greenhouse gas emissions, and within the building sector. And within the building sector, nearly 72% of the CO2 emissions are due operational carbon demand in the building sector. To meet the goals set out in the Paris Accord, we need to halve carbon emissions by the end of this decade: an aggressive and ambitious timeline. It has become an immediate and urgent need to design and construct buildings that are minimally intrusive in their impact on the environment and operate efficiently. Improvement of the building's energy performance must begin at the early design stage as the potential for sustainable intervention in the project's early phases is higher. The opportunity to improve building performance as the design progresses, is constantly reducing, while the cost of optimization is constantly increasing. One of the major design variables impacting the performance at the early phase is the building morphology/form and its associated variables. Sensible optioneering of the building form and its variables at the early stages of design - especially the conceptual design stage - can help improve the performance of the design early on in the design process at minimum cost. However, for effective early-phase design optimisation, there is a need to develop tools/methods that allow instantaneous evaluation of a large design sample space without much “lag” between the design and performance evaluation workflows. Also, in regular early phase sustainable design workflows, multiple stakeholders - like the designer, the performance analyst, etc. - are involved, which leads to a cognitive divide in the design process. For instance, while the architect designs the initial conceptual massing with a functional design intuition, the performance analysts suggest design interventions with a performance oriented design intuition. This cognitive divide leads to a dipartite design cognition in early phase design. The underlying premise of this research is to support the notion of blurring the dichotomy between conflicting design intuitions through exploring Human + Artificial Intelligence (AI) synergies and their underlying foundational technological requirements with respect to early phase design, to enable intuitive high performance design scenarios with a centralized design cognition where there are no other stakeholders other than the designer and the design environment. With the paradigm shift of other industries towards Machine Learning (ML) and AI, recent research advances in ML and the increasing availability of Big-data in the AEC industry have bolstered multivariate problems like building performance evaluation to a great extent. Design optioneering with instantaneous performance feedback through a ML method called Surrogate Modelling is an upcoming and promising methodology, which delivers feedback based on knowledge through available data, rather than simulation. Also, paradigms in Concurrent Human Machine Interaction (HMI) have explored avenues of augmenting the architectural design process through data-driven approaches. The type of data (input-output pairs) forming the basis for these paradigms depend on the target problem at hand. For early phase performance optimization - especially energy use optimization - Building Energy Model (BEM) forms the primary data. The US Department of Energy defines BEM as a versatile, multipurpose tool that is used in new building and retrofit design, code compliance, green certification, qualification for tax credits and utility incentives, and even real-time building control. BEM is also used in large-scale analyses to develop building energy-efficiency codes and inform policy decisions. A comprehensive BEM dataset that is large and accessible can support the growing interest in ML and HMI research in the high performance building design field. However, common techniques to acquire large BEM datasets like manual 3D energy modeling and simulation, 3D scanning, etc. are very tedious and time consuming. Also, existing datasets are usually incomplete, inconsistent and very difficult to access, owing to data privacy and open-access issues. And lastly, these datasets are fixed, static and cant be easily reproduced for different use case scenarios, nor scalable to adapt to needs. This research aims to tackle this problem by introducing a novel framework to generate custom problem-specific synthetic BEM dataset that generates a user defined amount of context specific 3D early phase building geometries and their associated BEM models, that is suitable for ML and HMI research. Synthetic datasets generated with this framework offer flexibility and customization in the generation process, making these datasets modular, reproducible and scalable. The framework uses the concepts of Generative Design, Geometry Manipulation, Simulation and Computational Automation to build the dataset. The dataset is qualified through the concepts of Preservationism and Sustainability, which are discussed in further sections of this article. The research identifies the importance and multifacetedness of the impact of building geometry on its performance, and demonstrates the application of generated synthetic BEM datasets by developing an enactive, conversational design environment that allows the designer to make real-time sustainable design decisions based on instantaneous machine feedback, for an intuitive, centralized design intuition.

Stakeholders such as urban planners and energy policymakers use building energy performance modeling and analysis to develop strategic sustainable energy plans with the aim of reducing energy consumption and emissions from the built environment. However, inconsistent energy data and the lack of scalable building models create a gap between building energy modeling and traditional planning practices. An alternative approach is to conduct a large-scale energy usage survey, which is time-consuming. Similarly, existing studies rely on traditional machine learning or statistical approaches for calculating large-scale energy performance. This paper proposes a solution that employs a data-driven machine learning approach to predict the energy performance of urban residential buildings, using both ensemble-based machine learning and end-use demand segregation methods. The proposed methodology consists of five steps: data collection, archetype development, physics-based parametric modeling, machine learning modeling, and urban building energy performance analysis. The devised methodology is tested on the Irish residential building stock and generates a synthetic building dataset of one million buildings through the parametric modeling of 19 identified vital variables for four residential building archetypes. As a part of the machine learning modeling process, the study implemented an end-use demand segregation method, including heating, lighting, equipment, photovoltaic, and hot water, to predict the energy performance of buildings at an urban scale. Furthermore, the model's performance is enhanced by employing an ensemble-based machine learning approach, achieving 91% accuracy compared to the traditional approach's 76%. Accurate prediction of building energy performance enables stakeholders, including energy policymakers and urban planners, to make informed decisions when planning large-scale retrofit measures.

On the relation between architectural considerations and heating energy performance of Turkish residential buildings in Izmir

As the world's population, urban infrastructure, and technological capabilities continue to expand rapidly, so does the demand for energy. This paper aims to emphasize the importance of prioritizing energy efficiency in new buildings and enhancing the energy performance of existing structures. This study reviews various machine learning (ML) models and their applications in building energy forecasting, comparing and contrasting their effectiveness. In recent years, ML approaches, particularly Artificial Neural Networks (ANNs), have been proposed for predicting energy consumption and performance in buildings. This paper discusses these models in the context of building energy forecasting. Furthermore, it explores the application of digital twins beyond the construction industry, highlighting their potential benefits in asset lifecycle management and optimization. By providing a comprehensive reviews of ML models and exploring the potential of digital twins, this research contributes to developing effective strategies for reducing energy consumption in the building sector.

Data-driven estimation of building energy consumption and GHG emissions using explainable artificial intelligence

PurposeThe purpose of this study is to quantify the energy heating performance of apartment buildings in Kosovo built after 2003 and compare it against the energy heating performance of buildings in member states of EU and selected European countries.Design/methodology/approachThis paper takes a case study approach focussed on the assessment of the heating energy performance of the building. This approach facilitated a detailed calculation of the selected materials’ energy performance used in a representative building structure in Kosovo comparing with passive buildings standard and energy heating performance of buildings in member states of EU and selected European countries.FindingsResults of quantitative research find that the energy heating performance of apartment buildings in Kosovo built after 2003 is far higher than that of passive buildings standard and is better than the average annual energy heating performance of apartment buildings in member states of the EU and selected European countries.Research limitations/implicationsThe research provides new knowledge regarding energy heating performance in new residential buildings in Kosovo and compares the findings with earlier research and energy consumption in other selected European countries. The research provides great benefits for researchers and practitioners working in the field of energy management as it compares the energy performance of residential buildings across Europe.Originality/valueThis paper provides a perspective on investigating the energy performance of a building structure of a residential apartment building in Prishtina, Kosovo. By unveiling the level of energy consumption of a residential apartment building in Kosovo representative of the new construction period can help the facility managers to acknowledge the standards they must achieve to refurbish the old building stock to achieve at least the same standard as the buildings in the new construction period.

National building renovation plans are a central element of the newly approved Energy Performance of Buildings Directive (European Commission, 2024). This plan will offer a comprehensive overview of the energy and environmental performance of both residential and non-residential buildings. To effectively map the energy status of the urban building stocks, it is useful to analyse building typologies that represent various climatic zones, building uses, and construction periods.Urban Building Energy Models (UBEMs) require vast amounts of data, which are often limited and inaccurate. Privacy policies significantly restrict data availability, impacting access to building information on a city-wide scale (HosseiniHaghighi et al., 2022; Johari et al., 2023). To address data uncertainty in large-scale energy analysis, collected data are used to create building archetypes (BAs). The BA approach strikes a balance between reducing complexity and enhancing the model’s accuracy in energy analysis. The energy performance certificates (EPCs) represent a core source of information to bridge the data uncertainty in large-scale analyses.Given the significant data uncertainty associated with generating BAs, deeper analysis of the effects of key input data deviations on energy performance assessment is necessary. This study begins with the classification of data in UBEMs, aiming to identify the essential inputs required for large-scale urban energy analysis. It then reviews and categorises existing Italian databases that can be used to reduce the high uncertainty of input data.The study involves generating probabilistic BAs based on EPCs from the Aosta Valley Region (Italy), which serves as a case study. A local large-scale sensitivity analysis was conducted, varying the thermo-physical parameters of the building fabric and the window-to-wall ratio of residential buildings in Aosta one-at-a-time. This analysis demonstrates how changes in the statistical ranges of inputs, in particular the performance of opaque building envelope components, affect the assessment of building energy needs.- European Commission. 2024. “Directive (EU) 2024/1275 of the European Parliament and of the Council of 24 April 2024 on the Energy performance of buildings (recast)”, Official Journal of the European Union, 24 April 2024.- HosseiniHaghighi, S., de Uribarri, P. M. Á., Padsala, R., and Eicker, U. 2022. “Characterizing and structuring urban GIS data for housing stock energy modelling and retrofitting”. Energy and Buildings 256. doi:10.1016/j.enbuild.2021.111706.- Johari, F., Shadram, F., and Widén, J. 2023. "Urban building energy modeling from geo-referenced energy performance certificate data: Development, calibration, and validation". Sustainable Cities and Society 96. doi:10.1016/j.scs.2023.104664.

The existing buildings stock in European countries accounts for over 40% of final energy consumption in the European Union (EU) member states. Consequently, an increase of building energy performance can constitute an important instrument in the efforts to alleviate the EU energy import dependency and comply with the Kyoto Protocol to reduce carbon dioxide emissions. This is also in accordance to the European Directive on the Energy Performance of Buildings (EPBD), which is currently under implementation in all EU member states. This paper presents an overview of EPBD and ongoing national activities, and focus on building energy performance assessment methodologies, in line with the EPBD. These methods and software can be used to perform building energy audits and assess buildings in a uniform way, perform demand and savings calculations, provide owners with specific advice for measures that can improve energy performance, and issue an Energy Performance Certificate (EPC) for existing buildings. Another ongoing European project is developing a common database structure and an evaluation scheme, which is being used to collect, process and evaluate data from 12 European countries. The results will constitute a good basis for the implementation of harmonized monitoring systems in the building sector on regional and national level.

It is widely accepted that the concentration of people living in high-density city centers offers greater operational energy efficiency and lower greenhouse gas emissions than lower-density expanded suburbs. The prevailing assumption is that lower-density suburbs are dominated by larger low-rise buildings that have higher building energy use requirements and greater per-person automobile travel requirements than high-density city centers dominated by medium- and high-rise buildings located in close proximity to a variety of public transit systems. However, very few studies to date have utilized empirical data at an individual household scale to evaluate differences in the operational energy (OE) footprints for both building and transportation energy end-uses between high-rise urban and low-rise suburban households. Therefore, this work collects empirical data on building and transportation OE consumption by individuals and households living in two economically similar groups: existing high-rise residential buildings in downtown Chicago, IL, USA and existing low-rise residential buildings in suburban Oak Park, IL, USA. Data were collected from over 500 households via an online survey. We considered the following components of residential living: (1) building OE (BOE), which includes electricity and/or natural gas use for all building energy end-uses; and (2) transportation OE (TOE), which includes the OE for multiple modes of transportation (i.e., automobile, bus, subway, regional train, etc.) based on average travel behavior in each location, as well as the OE for supporting transportation infrastructure. We estimate that downtown high-rise living in this sample of residences in Chicago, IL accounts for approximately 427 GJ of primary OE per household per year, on average, which was 14% lower than the average for suburban low-rise living in the Oak Park, IL homes (499 GJ per household per year). However, on a per-person basis, downtown high-rise living accounts for approximately 246 GJ of primary OE per person per year, which was approximately 61% higher than suburban low-rise living (153 GJ per person per year). In both building types, building OE was the single largest contributor to total OE use. This study accurately captured the energy requirements associated with realistic behaviors and lifestyles of occupants of both low-rise suburban and high-rise urban households, and found that building OE dominates the total OE, which suggests that efforts to reduce building OE should be given high priority in building design and management as well as urban planning.

With the increased awareness of energy consumption as well as the environmental impact of building operations, architects, designers and planners are required to place more consideration on sustainability and energy performance of the building. To ensure most of those considerations are reflected in the building performance, critical design decisions should be made by key stakeholders early during the design development stage. The application of BIM during building energy simulations has profoundly improved the energy analysis process and thus this approach has gained momentum. However, despite rapid advances in BIM-based processes, the question still remains how ordinary building stakeholders can perform energy performance analysis, which has previously been conducted predominantly by professionals, to maximize energy efficient building performance. To address this issue, we identified two leading building performance analysis software programs, Energy Plus and IES (IES ), and compared their effectiveness and suitability as BIM-based energy simulation tools. To facilitate this study, we examined a case study on Building Performance Model (BPM) of a single story building with one door, multiple windows on each wall, a slab and a roof. We focused particularly on building energy performance by differing building orientation and window sizes and compared how effectively these two software programs analyzed the performance. We also looked at typical decision-making processes implementing building energy simulation program during the early design stages in the U.S. Finally, conclusions were drawn as to how to conduct BIM-based building energy performance evaluations more efficiently. Suggestions for further avenues of research are also made.

With the increased awareness of energy consumption as well as the environmental impact of building operations, architects, designers and planners are required to place more consideration on sustainability and energy performance of the building. To ensure most of those considerations are reflected in the building performance, critical design decisions should be made by key stakeholders early during the design development stage. The application of BIM during building energy simulations has profoundly improved the energy analysis process and thus this approach has gained momentum. However, despite rapid advances in BIM-based processes, the question still remains how ordinary building stakeholders can perform energy performance analysis, which has previously been conducted predominantly by professionals, to maximize energy efficient building performance. To address this issue, we identified two leading building performance analysis software programs, Energy Plus and IES (IES ), and compared their effectiveness and suitability as BIM-based energy simulation tools. To facilitate this study, we examined a case study on Building Performance Model (BPM) of a single story building with one door, multiple windows on each wall, a slab and a roof. We focused particularly on building energy performance by differing building orientation and window sizes and compared how effectively these two software programs analyzed the performance. We also looked at typical decision-making processes implementing building energy simulation program during the early design stages in the U.S. Finally, conclusions were drawn as to how to conduct BIM-based building energy performance evaluations more efficiently. Suggestions for further avenues of research are also made.

<p>In 2008, the European Energy Performance of Buildings Directive (EPBD) facilitated the introduction of two mandatory energy assessment methods in the UK. Energy Performance Certificates (EPCs) reveal the modelled energy performance of buildings when they are constructed, sold or let based on their intrinsic energy attributes, whereas Display Energy Certificates (DECs) reveal operational energy performance in a subset of buildings that is operated by the public sector, based on annual energy consumption data.</p>\n\n<p>EPCs were conceived as a marketing mechanism for property market participants and they have been used in studies that have sought to investigate the links between energy performance and financial performance of buildings. Past studies have investigated the relationship between modelled and operational energy performance measurement in buildings and have found mismatches between both. </p>\n\n<p>This study will investigate the link between the modelled and actual energy performance in office buildings that are occupied by the public sector. The study uses detailed EPC and DEC data from the Department for Communities and Local Government. A comprehensive benchmarking analysis of these ratings establishes the extent to which both align and differ across the same units. The EPC and DEC data is also matched to data on building attributes from CoStar UK to investigate the relationship between energy performance and building features such as age and building quality, which have been commonly used as control variables in past hedonic pricing studies. This study will also look at the magnitude of observed changes in operational energy performance over time, to investigate whether energy performance assessment leads to energy performance improvement.</p>\n\n<p>These findings will provide further insights into the effects and impacts of the introduction of energy certification for buildings. The further aim of this study is to develop a building typology based on commonly shared building and energy performance attributes.</p>

In 2008, the European Energy Performance of Buildings Directive (EPBD) facilitated the introduction of two mandatory energy assessment methods in the UK. Energy Performance Certificates (EPCs) reveal the modelled energy performance of buildings when they are constructed, sold or let based on their intrinsic energy attributes, whereas Display Energy Certificates (DECs) reveal operational energy performance in a subset of buildings that is operated by the public sector, based on annual energy consumption data.EPCs were conceived as a marketing mechanism for property market participants and they have been used in studies that have sought to investigate the links between energy performance and financial performance of buildings. Yet they are based on modelled energy performance and the ratings that they express are hypothetical, whereas DECs are based on actual energy consumption figures. Furthermore, EPCs are valid for 10 years, whereas DECs need to be renewed annually.This study will investigate energy performance patterns as recorded in DEC certificates in existing office buildings over time. The study uses detailed DEC data for commercial office buildings from the Department for Communities and Local Government. This dataset is matched to data on building attributes from CoStar UK to investigate the relationship between energy performance and building features such as age and building quality. This study models the magnitude of observed changes in operational energy performance in existing buildings, to investigate how operational energy performance assessment can be used to track energy performance improvements over time, and reveal how different control variables may impact on recorded changes. These findings will provide further insights into the effects and impacts of the introduction of energy certification for buildings. The further aim of this study is to develop a building typology based on commonly shared building and energy performance attributes.

AbstrctThere is a growing interest in sustainable design in the building industry to reduce energy consumption and minimize adverse environmental impacts of buildings. The strategies for sustainable design are as follows: 1) reducing the size of the building′s equipment system and saving energy through an optimal design; 2) maximizing natural energy use through a passive solar heating system; and 3) utilizing an active system through applications of high-performance heating, ventilation, and air conditioning (HVAC) and lighting systems, installation of new and renewable energy facilities, and so on. It is vital to evaluate and compare the energy efficiencies of design alternatives at an early design stage, and hence, to improve the energy performance of the final building, as design elements determined at an early phase in the architectural design process greatly influence the energy performance of the building itself. Further, costs increase over time with the number of design changes made. In the course of this research, the KLT (Korean lighting and thermal energy) method was revised and developed based on the lighting and thermal energy (LT) method, adjusting for South Korea′s climate and architectural regulations, which can be used to assess the energy performance of buildings. This study was conducted to determine the process of selecting optimal design alternatives to maximize building energy performance at an early stage in the process.