Machine Learning Techniques Focusing on the Energy Performance of Buildings: A Dimensions and Methods Analysis

Maria Anastasiadou,Miguel Sales Dias,Vítor Santos

doi:10.3390/buildings12010028

Maria Anastasiadou, Miguel Sales Dias + Show 1 more

Open Access

https://doi.org/10.3390/buildings12010028

Copy DOI

Abstract

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 measures. The main contribution of our study is a conceptual and theoretical framework applicable in the analysis of the energy performance of buildings with intelligent computational methods. With our framework, the reader can understand which approaches are most used and more appropriate for analyzing the energy performance of different types of buildings, discussing the dimensions that are better used in such approaches.

Highlights

Considering that buildings account for 40% of the primary energy consumption (EC) in the European Union [1], reducing the EC of buildings has become a necessity
In the current state of the art, data science and machine learning are available to analyze, predict and improve energy efficiency (EE) in buildings in meaningful ways. Such computer science approaches can be used to forecast and minimize energy consumption, design energy-efficient buildings, define strategies for mitigating impacts on the environment and climate, and predict and propose useful and cost-effective retrofit measures to increase the EE of buildings to provide a comfortable indoor living environment [9,10]
This paper proposes a conceptual and theoretical framework applicable in the analysis of literature papers that tackle the problem of the energy performance of buildings (EPB) with machine learning or statistical methods

Summary

Introduction

Considering that buildings account for 40% of the primary energy consumption (EC) in the European Union [1], reducing the EC of buildings has become a necessity. Union, considering the increasing urbanization and climate change trends, defined the objective to reduce EC by 32.5% until 2030, from the baseline year of 2007, as a key priority in the EU’s strategy and Green deal [2] to increase EE and decrease the energy performance (EP) of existing buildings [2,3,4]. This goal is aligned with the United Nations’ seventh. By measuring, monitoring, and improving the EE in buildings, we can reduce the amount of energy consumed while maintaining or even enhancing the quality of services provided by those buildings, a “double the global rate of improvement in EE”—SDG7.8 [5,11]

Objectives

Methods

Results

Conclusion