Modelling an Artificial Intelligence-based Energy Management for House/Mini-grid in Sub-Saharan Africa: A Case Study of Nigeria

Abstract

It is no longer a news up to 600 million people living in the Sub-Saharan Africa (SSA) do not have access to electricity and 80% of them are located in the rural areas of the continent. The need for energy conservation especially electricity is of crucial importance as it is an economic solution to the problem of energy shortage and atmospheric carbon reduction. Buildings has been identified among the top largest world energy consumption accounting for up to 40% and in order to maximize energy conservation, there is need to put in place an effective energy management strategy. The role of Artificial technology has also been displayed by researchers in the promotion of energy management. Most of past literatures in the line of energy management strategies proposed various energy management model based on smart grid and smart meter technology, demand side management, home energy management schemes and management based on Artificial Intelligence. However, majority of these proposed models are focused on the Urban regions mostly in developed countries. There are very few literatures focusing on energy management model strategy based on Artificial intelligence. Hence there is a knowledge gap. The aim of this work is to model an Artificial intelligence-based energy management for households and mini-grid in SSA, specifically Nigeria. Genetic algorithm was used on smart meter-like data to optimize the energy consumption of households (i.e. low, middle and high income earners) for 24 hours in both weekday and weekend. To achieve this aim, we determine the typical load profile of a mini-grid setting (for household and commercial load profile), develop a simulation of smart meter-like data and develop an energy management system to optimize electricity consumption during a week day and weekend in a household. Based on experimental results; the energy (and consumption) saved during week day for high, middleand low-income earners are 16.74%, 9.9% and 32.55%, respectively. Similarly, the corresponding cost and consumption saved during weekends for high, middle- and low-income earners are 6.11%, 19.39% and 32.20%, respectively. The smart system algorithm will not only help the consumer to reduce electricity cost by reducing consumption but also help to reduce demand at the grid level.