Investigating the Drivers of Grid Electricity Demand in Nigeria: Harnessing the Power of Machine Learning and Artificial Intelligence Algorithms

Author:

Olubusoye Olusanya E.1,Emmanuel Precious M.1,Nwobi Lucy D.1,Daramola Abayomi1,Ajulo Kayode D.1,Adeyanju Omosalewa T.1,Oyebade Bolu J.1,Adejumo Olabode1,Akintande Olalekan J.1

Affiliation:

1. University of Ibadan

Abstract

Abstract The electricity situation in Nigeria is epileptic, and several households rely on alternative sources of electricity to power homes and businesses. In an urban area of the country, four (4) out of every six (6) households operate an alternative electricity supply. On the other hand, individuals in rural areas can hardly afford alternative electricity sources. More so, urban individuals use mostly alternative sources that are non-renewable (such as generators), thereby causing environmental degradation and promoting Climate-related issues. The direct impact of these alternative electricity sources has led to the death of entire families due to carbon monoxide breathing by household occupants/family and caused detrimental health issues to nearby households on several occasions. To address electricity instability issues and promote a sustainable and reliable green energy future, it is worthwhile to identify the factors influencing the grid electricity demand and develop efficient models for identifying these factors and their influence on grid electricity demand. The study finds that classical algorithms are ultimately inefficient due to the multicollinearity component of the relevant features. Consequently, the study harnesses the power of Blackbox and Glassbox algorithms - Deep Neural Network (DNN) and Multivariate Adaptive Regression Spline (MARS), respectively, to investigate these factors. The two learning algorithms agreed that Nigeria's electricity demand is majorly driven by Macroeconomic and Climatology variables – with Rural population, Temperature and GDP per capita being the most relevant drivers of electricity demand in Nigeria. Given the relevance of the GDP per capita, the result implies that the discrepancies in the socioeconomic characteristics of households or individuals in Urban to Rural played a major role in the electricity demand in Nigeria. Hence, the study concludes that by addressing only Electricity problems, Nigeria will achieve fifty-three (53) percent of the global SDG agenda and greater economic development.

Publisher

Research Square Platform LLC

Reference39 articles.

1. Performance evaluation of the prospects and challenges of effective power generation and distribution in Nigeria;Adoghe AU;Science Direct,2023

2. Recurrent Neural Network Model for Forecasting Electricity Demand in Nigeria;Abdusalam KA;Journal of Engineering Research,2016

3. Adedeji, P., Madushele, N. and Akinlabi, S., 2018, September. Adaptive Neuro-fuzzy Inference System (ANFIS) for a multi-campus institution energy consumption forecast in South Africa. In Proceedings of the International Conference on Industrial Engineering and Operations Management (pp. 950–958).

4. A deep learning model for electricity demand forecasting based on a tropical data;Adewuyi SA;Applied Computer Science,2020

5. Modelling and forecasting hourly electricity demand in West African countries;Adeoye O;Applied Energy,2019

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3