Machine Learning for Energy Management in Buildings: A Systematic Review on Real-World Applications

This paper reviews the work in the areas of machine learning applications for energy management in smart buildings, 5G technology’s role in smart energy management, and the use of machine learning algorithms in microgrid energy management systems. The first area focuses on the adaptability of building-integrated energy systems to unpredictable changes through AI-initiated learning processes and digital twins. The second area explores the impact of 5G technology on smart buildings, particularly in Singapore, emphasizing its role in facilitating high-class services and efficient functionalities. The third area delves into the application of various machine learning algorithms, such as supervised and unsupervised learning, in managing and monitoring microgrids. These broad areas collectively offer a comprehensive understanding of how machine learning can revolutionize energy management systems in smart buildings, making them more efficient, adaptable, and sustainable.

The present study provides potentially interesting for researchers in the novel integrated method of the fuzzy set qualitative comparative analysis (fsQCA) and digital design methodology to apply in the field of energy and sustainability of socio-technical systems (STSs) used in the building energy management (BEM). The research sample is the air-conditioning system (ACS). The social and technical aspects of ACS are satisfaction and energy consumption. Ten sustainability criteria are applied to assess and monitor the fields of energy consumption and satisfaction with the aim of achieving sustainable consumption in the use of ACSs for BEM. This action can be carried out by nesting a circuit in a microprocessor and microcontroller for use in a building energy management system (BEMS). According to, the circuit obtained from the assessment of the study can be practically implemented and assembled in a BEMS as online. This work can define a sustainability-based BEMS.

Objective: This study aims to identify key characteristics of smart buildings that enhance sustainability and energy efficiency in educational institutions. By leveraging artificial intelligence, particularly machine learning, the study seeks to optimize energy consumption and improve learning environments. Theoretical Framework: Smart buildings integrate digital and communication technologies to enhance educational infrastructure. Despite their benefits, adoption in developing countries remains limited. This study builds on theories of sustainable energy management and computational optimization, emphasizing machine learning’s role in predictive energy modeling. Method: A machine learning model was developed to predict energy consumption in educational institutions. The k-nearest neighbor (k-NN) algorithm was applied using open-access data from the University of Durrës Building Energy Management System. Model validation was conducted through comparative analysis, assessing prediction accuracy and energy-saving potential. Results and Discussion: The model achieved an average relative error of 18.26%, confirming its predictive capability. Smart features enabled energy savings between 44.8% and 58.3%, depending on analyzed spaces. Additionally, an AI-powered interactive dashboard was proposed for real-time energy monitoring and optimization, aiding facility managers. Research Implications: This study underscores the role of smart buildings in promoting sustainability and reducing operational costs. AI-driven energy management systems can significantly enhance institutional sustainability practices. Originality/Value: This research demonstrates the effectiveness of machine learning in energy consumption prediction and proposes an AI-based solution for sustainable energy management in educational buildings. The findings provide valuable insights into integrating smart technologies for sustainability and efficiency.

Market and behavior driven predictive energy management for residential buildings

Building and district energy management increasingly requires advanced control strategies. For several tasks that arise, Machine Learning (ML) and Artificial Intelligence (AI) have found widespread application, but often the interaction with existing elements and the integration into the productive system are done in a rather unsystematic way. Thus, one faces problems concerning interface design, data requirements and reliability/trustworthiness. In order to improve this situation, we outline the general control structure of such systems, review key concepts from ML/AI and discuss approaches for conventional high-level control. This is followed by a literature review on ML/AI applications for energy management. Based on the systematics and the literature findings, we establish a taxonomy for the integration of ML/AI methods for energy management, which leads as a main result to a comprehensive guide for such applications, taking into account both characteristics of the method and the affected elements of the system. This is supplemented by a quick guideline for choice of an appropriate method and for subsequent evaluation. The target readership are control engineers who would like to get a systematic overview of ways to integrate ML/AI methods in their work, and data scientists who want to get a better understanding for main tasks and challenges for applying their tools for control tasks.

A linear programming based heuristic algorithm for charge and discharge scheduling of electric vehicles in a building energy management system

Model predictive control for energy management in buildings Part 2: Distributed Model Predictive Control

Building energy management is recognized as a useful technique to enhance building automation and facilitate demand response. This paper proposes a simple and useful building energy management system (BEMS) that simultaneously minimizes the user's electricity bill while assistants the utility to reduce the Peak-to-Average Ratio (PAR) of the regional load. The working mechanism of the proposed BEMS is based on a simple observation: the PAR reduction of an area can be achieved by reducing the PAR of individual buildings. The system consists of two stages: (1) in the first stage, the appliances are temporarily scheduled to obtain the minimum electricity cost of the unit; and (2) in the second stage, the appliances are re-scheduled with the objective to minimize the unit's PAR, while ensuring the building's electricity cost does not exceed the optimal cost determined in the first stage. Our numerical results show that by applying the proposed BEMS to individual units, the overall PAR of a group of buildings can be effectively reduced.

This article is important for professionals related to the world of infrastructure and building, because it provides what is necessary to understand energy management in buildings.It is also important for the technological impact it provides to users [22].This research analyzes the energy management in smart buildings in Peru through methodologies such as the review of articles published in the Scopus database using the VOSviewer software as a data analysis tool.This analysis has a period from the XXI century to the year 2021.Graphs related to the research of the database were obtained where the words "intelligent building" and "energy management" are highlighted with greater congruence and mentioned in these journals.Figures were obtained worldwide and in Peru.The results allowed us to understand that smart buildings are becoming more important in developed countries and opening new markets since there are unmet needs of users.In the analysis of Peru, the opposite happens since there are minimal scientific contributions related to the subject.They still do not have intelligent buildings 100% but there are some buildings where energy management is carried out in minimum proportion.An energy management software related to big data called smart building was also found, which has the objective of managing energy consumption in each building to reduce the carbon footprint emitted by buildings.This research is intended to guide you to start with in-depth research and implementation projects for engineers and architects interested in developing and managing intelligent edifications.Nowadays, the importance of studying this topic is to take care of the environment as social and professional responsibility.For this reason, the professionals should opt for better strategies for energy management thus positively impact Peru.

Coordination of vehicle-to-home and renewable capacity resources for energy management in resilience and self-healing building

The oil and gas industry faces constant pressure to enhance operational efficiency, reduce costs, and minimize equipment downtime. Machine learning (ML) applications in predictive maintenance have emerged as a transformative approach to achieving these goals. This review explores the role of machine learning in predictive maintenance, highlighting its potential to revolutionize maintenance strategies and improve asset management across the industry. Predictive maintenance leverages advanced algorithms to analyze historical and real-time data from equipment and sensors, enabling the identification of patterns and anomalies that precede equipment failures. By utilizing techniques such as supervised learning, unsupervised learning, and reinforcement learning, organizations can forecast equipment malfunctions and schedule maintenance activities proactively, thereby reducing unexpected downtimes and extending asset lifecycles. The implementation of ML in predictive maintenance provides several key benefits. Firstly, it enhances operational efficiency by optimizing maintenance schedules and minimizing unplanned outages, which are critical in the capital-intensive oil and gas sector. Secondly, it enables cost savings through more efficient resource allocation and reduced labor costs associated with reactive maintenance strategies. Thirdly, machine learning algorithms can continuously learn from new data, refining their predictive capabilities and improving accuracy over time. Several case studies illustrate the successful application of machine learning in predictive maintenance within the oil and gas industry. For example, ML models have been employed to predict pump failures, optimize drilling operations, and improve pipeline integrity monitoring. These applications not only lead to significant financial savings but also enhance safety by reducing the risk of catastrophic failures. In conclusion, machine learning applications in predictive maintenance represent a crucial advancement for the oil and gas industry. By harnessing the power of data-driven insights, organizations can enhance operational efficiency, reduce costs, and ultimately drive sustainable growth. This review emphasizes the transformative potential of machine learning in predictive maintenance, establishing it as a key strategy for success in the ever-evolving oil and gas landscape.

This paper presents the structural review of a building energy management system (BEMS) based on facility sensor networks for each facility in the target building. To make BEMS works, sensors have to be installed for each facility in the target building. Sensors for each facility are connected to the data acquisition device (DAQ) and construct a sensor network. Each DAQ in a facility sensor network is connected to the gateway. Building working information is transferred from the gateway to the BEMS server. The BEMS uses collected data to analyze the building energy usage status and facility efficiencies. Several methods are possible for analyzing building working status based on status data. The easiest method is comparing realtime data to the spec data. Comparing realtime data to the building simulation result is the hardest method. To make BEMS works efficiently, it is possible to design BEMS as a multi-engine based system.

A concept review of power line communication in building energy management systems for the small to medium sized non-domestic built environment

A hybrid decision support system for adaptive trading strategies: Combining a rule-based expert system with a deep reinforcement learning strategy

Actor-critic learning for optimal building energy management with phase change materials