Monitoring, Predicting, and Optimizing Energy Consumptions: A Goal Toward Global Sustainability

Introduction

A recent study by the International Energy Agency (2019) shows the continuous increase of the global energy consumption which, in 2018, nearly doubled the average of the growth between 2010 and 2017. This growth was driven by a demanding global economy and population needs such as the higher consumptions in the heating and cooling of buildings which, for instance, in 2012 accounted for around 50% of the consumptions in the European Union (EU Publications, 2016). This led to an increasing demand of energy with significant growth in the consumption of renewable and non-renewable energy sources, with CO₂ emission raising almost 2% between 2017 and 2018, as a consequence. In 2018, electricity demand was accountable for more than 50% of the energy needs (International Energy Agency, 2019), contributing to two-thirds of emissions growth. Nevertheless, there are some good news as emissions declined for some of the most industrialized countries, such as Germany, Japan, Mexico, France, and the United Kingdom, and renewables incorporation increased by 4% in 2018, accounting for almost one-quarter of global’s energy demand growth.

Globally, more than 40% of total energy consumption and 30% of total CO₂ emissions are associated to buildings. In the European Union, buildings also represent 40% of total energy consumption and approximately 36% of greenhouse gas emissions (Ahmad, Mourshed, Mundow, Sisinni, & Rezgui, 2016; Ahmad, Mourshed, & Rezgui, 2017), being urgent the contraction or, at least, a stabilization of those values. Some measures to reduce the energy needs must therefore be taken, supported on consuming politics, buildings’ improvements, consumers’ awareness, or intelligent buildings promotion. For instance, artificial intelligence techniques can be used to analyze the history of energy consumption of buildings and to predict its future consumption levels. For instance, the integration of predictive power consumption models is one essential component in energy control and operating strategies (Pinto et al., 2019). This can contribute to the minimization of the energy costs of buildings, using Energy Management Systems (EMS) capable of scheduling the loads according with the generation levels or tariff rates. It can also be used in the development of decision support systems, that inform users about the best solutions to deal with a certain consumption pattern, and/or the implementation of automated systems for the detection of excessive consumption levels. All these solutions require the development and/or usage of prediction algorithms but, accurate forecasting of consumption is a challenge and a very complex task, not only because these are typically non-linear systems, but also correlate to seasonal variations and atmospheric conditions, while being dependent on users’ habits.

In many countries, renewable energy production already represents an important percentage of the total energy that is generated in electrical grids. In order to reach higher levels of integration, demand side management measures are required. In fact, different from the traditional electrical grids, where at any given instant the production is adjusted to meet the demand, when using renewable energy sources, the demand must be adapted in accordance with the generation levels, since these cannot be controlled. EMS can have here a major role to monitor both the generation and consumption patterns, and to control electrical appliances, alleviating users from the burden of individual control of each appliance. Therefore, a platform for the monitoring, prediction, and optimization of consumptions in buildings is an asset that can contribute to the improvement of today’s reality.