Machine Learning Practices During the Operational Phase of Buildings: A Critical Review (original) (raw)
Related papers
State-of-the-Art on Research and Applications of Machine Learning in the Building Life Cycle
Energy and Buildings
Fueled by big data, powerful and affordable computing resources, and advanced algorithms, machine learning has been explored and applied to buildings research for the past decades and has demonstrated its potential to enhance building performance. This study systematically surveyed how machine learning has been applied at different stages of building life cycle. By conducting a literature search on the Web of Knowledge platform, we found 9579 papers in this field and selected 153 papers for an in-depth review. The number of published papers is increasing year by year, with a focus on building design, operation, and control. However, no study was found using machine learning in building commissioning. There are successful pilot studies on fault detection and diagnosis of HVAC equipment and systems, load prediction, energy baseline estimate, load shape clustering, occupancy prediction, and learning occupant behaviors and energy use patterns. None of the existing studies were adopted broadly by the building industry, due to common challenges including (1) lack of large scale labeled data to train and validate the model, (2) lack of model transferability, which limits a model trained with one data-rich building to be used in another building with limited data, (3) lack of strong justification of costs and benefits of deploying machine learning, and (4) the performance might not be reliable and robust for the stated goals, as the method might work for some buildings but could not be generalized to others. Findings from the study can inform future machine learning research to improve occupant comfort, energy efficiency, demand flexibility, and resilience of buildings, as well as to inspire young researchers in the field to explore multidisciplinary approaches that integrate building science, computing science, data science, and social science.
Machine Learning for Smart and Energy-Efficient Buildings
arXiv (Cornell University), 2022
Energy consumption in buildings, both residential and commercial, accounts for approximately 40% of all energy usage in the U.S., and similar numbers are being reported from countries around the world. This significant amount of energy is used to maintain a comfortable, secure, and productive environment for the occupants. So, it is crucial that the energy consumption in buildings must be optimized, all the while maintaining satisfactory levels of occupant comfort, health, and safety. Recently, Machine Learning has been proven to be an invaluable tool in deriving important insights from data and optimizing various systems. In this work, we review the ways in which machine learning has been leveraged to make buildings smart and energy-efficient. For the convenience of readers, we provide a brief introduction of several machine learning paradigms and the components and functioning of each smart building system we cover. Finally, we discuss challenges faced while implementing machine learning algorithms in smart buildings and provide future avenues for research at the intersection of smart buildings and machine learning.
Energy and Buildings, 2019
After decades of evolution and improvements, Artificial Intelligence (AI) is now taking root in our daily lives, and is starting to profoundly influence the fields of architecture and sustainability. The applications of AI to sustainable architecture include energy-efficient building design, forecasting and minimizing energy consumption, strategizing for mitigating impacts on environment and climate, and enhancements in the safety and comfort of the living environment. Due to the significant increases in internet speed and accessibility and the drops in computer prices and data storage costs in recent years, Big Data (BD) nowadays plays an important supplementary role to AI. Algorithms and computer codes have been developed for data mining and analysis. BD rejuvenates AI methods and applications in many areas, including sustainable architecture. The present paper starts with an introduction to AI history and techniques. This is followed by a discussion on how AI and BD can be used to design and operate energy-efficient commercial buildings and residential houses, followed by a review of recent applications of AI and BD to energy-efficient buildings with an emphasis on the use of machine learning (ML) and large databases. Future research topics are suggested at the end of this paper. It is reemphasized in the present paper that AI, when combined with BD, can tremendously increase the energy efficiency and cost effectiveness of buildings which are designed to provide occupants with a comfortable indoor living environment.
Improving Energy Efficiency in Buildings Using Machine Intelligence
Lecture Notes in Computer Science, 2009
Improving the detection of thermal insulation in buildings -which includes the development of models for heating and ventilation processes and fabric gain -could significantly increase building energy efficiency and substantially contribute to reductions in energy consumption and in the carbon footprints of domestic heating systems. Thermal insulation standards are now contractual obligations in new buildings, although poor energy efficiency is often a defining characteristic of buildings built before the introduction of those standards. Lighting, occupancy, set point temperature profiles, air conditioning and ventilation services all increase the complexity of measuring insulation efficiency. The identification of thermal insulation failure can help to reduce energy consumption in heating systems. Conventional methods can be greatly improved through the application of hybridized machine learning techniques to detect thermal insulation failures when a building is in operation. A three-step procedure is proposed in this paper that begins by considering the local building and heating system regulations as well as the specific features of the climate zone. Firstly, the dynamic thermal performance of different variables is specifically modelled, for each building type and climate zone. Secondly, Cooperative Maximum-Likelihood Hebbian Learning is used to extract the relevant features. Finally, neural projections and identification techniques are applied, in order to detect fluctuations in room temperatures and, in consequence, thermal insulation failures. The reliability of the proposed method is validated in three winter zone C cities in Spain. Although a great deal of further research remains to be done in this field, the proposed system is expected to outperform conventional methods described in Spanish building codes that are used to calculate energetic profiles in domestic and residential buildings.
An overview of machine learning applications for smart buildings
Sustainable Cities and Society
The efficiency, flexibility, and resilience of building-integrated energy systems are challenged by unpredicted changes in operational environments due to climate change and its consequences. On the other hand, the rapid evolution of artificial intelligence (AI) and machine learning (ML) has equipped buildings with an ability to learn. A lot of research has been dedicated to specific machine learning applications for specific phases of a building's life-cycle. The reviews commonly take a specific, technological perspective without a vision for the integration of smart technologies at the level of the whole system. Especially, there is a lack of discussion on the roles of autonomous AI agents and training environments for boosting the learning process in complex and abruptly changing operational environments. This review article discusses the learning ability of buildings with a system-level perspective and presents an overview of autonomous machine learning applications that make independent decisions for building energy management. We conclude that the buildings' adaptability to unpredicted changes can be enhanced at the system level through AI-initiated learning processes and by using digital twins as training environments. The greatest potential for energy efficiency improvement is achieved by integrating adaptability solutions at the timescales of HVAC control and electricity market participation. 'intelligent building' (IB) and 'smart building' (SB) (Al Dakheel et al. (2020); Wang et al. (2020))). A shift towards the implementation of artificial intelligence (AI) trained by machine learning algorithms is recognized as one of the major trends of development (Karpook, 2017). Given the complexities related to the operational environment, the machine learning techniques 'reinforcement learning (RL)' and its derivative 'deep reinforcement learning (DRL)' have been experienced useful for the autonomous control networks of buildings (Han et al., 2019). Quite a few review articles have been published with various perspectives on smart buildings. A quick look at the most relevant review articles in the field reveals that most of them focus on issues such as hardware technologies, monitoring, forecasting, modelling, building energy management, and applications of machine learning (Alawadi et al.
Machine Learning for Smart Building Applications: Review and Taxonomy
2018
The use of machine learning (ML) in smart building applications is reviewed in this paper. We split existing solutions into two main classes, occupant-centric vs. energy/devices centric. The first class groups solutions that use ML for aspects related to the occupants, including (1) occupancy estimation and identification, (2) activity recognition, and (3) estimating preferences and behavior. The second class groups solutions that use ML to estimate aspects related either to energy or devices. They are divided into three categories, (1) energy profiling and demand estimation, (2) appliances profiling and fault detection, and (3) inference on sensors. Solutions in each category are presented, discussed and compared, as well as open perspectives and research trends. Compared to related state-of-the-art survey papers, the contribution herein is to provide a comprehensive and holistic review from the ML perspectives rather than architectural and technical aspects of existing building ma...
Buildings, 2021
The problem of energy consumption and the importance of improving existing buildings’ energy performance are notorious. This work aims to contribute to this improvement by identifying the latest and most appropriate machine learning or statistical techniques, which analyze this problem by looking at large quantities of building energy performance certification data and other data sources. PRISMA, a well-established systematic literature review and meta-analysis method, was used to detect specific factors that influence the energy performance of buildings, resulting in an analysis of 35 papers published between 2016 and April 2021, creating a baseline for further inquiry. Through this systematic literature review and bibliometric analysis, machine learning and statistical approaches primarily based on building energy certification data were identified and analyzed in two groups: (1) automatic evaluation of buildings’ energy performance and, (2) prediction of energy-efficient retrofit...
Machine Learning, Big Data, And Smart Buildings: A Comprehensive Survey
2019
Future buildings will offer new convenience, comfort, and efficiency possibilities to their residents. Changes will occur to the way people live as technology involves into people's lives and information processing is fully integrated into their daily living activities and objects. The future expectation of smart buildings includes making the residents' experience as easy and comfortable as possible. The massive streaming data generated and captured by smart building appliances and devices contains valuable information that needs to be mined to facilitate timely actions and better decision making. Machine learning and big data analytics will undoubtedly play a critical role to enable the delivery of such smart services. In this paper, we survey the area of smart building with a special focus on the role of techniques from machine learning and big data analytics. This survey also reviews the current trends and challenges faced in the development of smart building services.
Developing energy efficient building design in machine learning
Isarc Proceedings, 2010
Building energy simulation programs have been developed, enhanced, and are in widespread use throughout the building construction community (Stumpf et al. 2009). Energy modeling programs provide users with key building performance indicators such as energy use and demand, temperature, humidity, and costs. As the A/E/C industry is embracing the technology of energy simulation programs, building designers are currently encountering a large amount of data generated during energy simulations. From our experience, even a simple energy modeling run generates hundreds of pages of data. Examples of building features simulated include the estimated energy costs in terms of building orientation, HVAC system, lighting efficiency and control, roof and wall insulation and construction, glazing type, water usage, day-lighting and so on. Such volumes of data are simply beyond human abilities to identify the best combination of building components (insulation, windows, doors, etc.) and systems (heating and cooling systems, ventilation, etc.) during the building design process. Evaluating building energy modeling outputs clearly overwhelms the traditional methods of data analysis such as spreadsheets and ad-hoc queries. This paper presents the analysis to develop the energy efficient solutions with the baseline and target energy estimations. Finally, energy efficient solutions are presented that enable the energy savings to be met in fifteen different climate zones in the United States.
Building energy management and data analytics
2015 International Symposium on Smart Electric Distribution Systems and Technologies (EDST), 2015
Energy efficiency in buildings depends on the way the building is operated. Therefore energy management is the key component for efficient operation. Data analysis of operation data helps to better understand the systems and detect faults and inefficiencies. The facility manager benefits from smart analysis that makes use of machine learning algorithms and innovative visualizations. This analysis is part of a bigger review of the current structure of building automation as it is used in today's buildings. The operation targets in energy efficiency are complex, ambiguous and contradictory: indoor comfort, energy efficiency, high availability and low costs cannot be met at the same time. In order to improve building operation, a novel model of automation is discussed. The foundation of this model is in cognitive automation, since each building is unique in its selection of energy sources, architecture, usage and location, which implies that the building's control system has to be adapted individually. This paper connects the data-driven analysis of operation data with a cognitive concept to be used for operating the energy systems in a building and regarding goals on how to optimally operate while considering constraints about the limits of operation, using the complex, dynamic data from building automation.