ABSTRACT
Electricity access is a challenge faced globally, especially in sub-Saharan Africa (SSA), where over half of the world's energy-poor population is located. Energy data provided by evolving smart metering (SM) technologies, applied to energy access frameworks such as Sustainable Energy Access Planning (SEAP), can enable the planning, implementation, and maintenance of energy systems for energy-poor areas. This study aims to analyse such applicability of SM to the SEAP framework, where household (HH) energy data based on an SSA country's criteria for energy access is studied and defined into an energy profile, which then serves as input to a smart metering experiment comprising a household's electrical load, a smart meter that measures electricity usage, and a Meter Data Management System (MDMS) software that remotely collects the data through General Packet Radio Service (GPRS) communications. An analysis is performed on the applicability of the acquired energy data to the SEAP framework and over three exercises that explore the analysis and calculation of energy access indicators, demand forecasting via machine learning (ML), and the optimization and cost analysis of energy systems. The household's measured energy data not only applies to all the assessments in the SEAP framework but also reveals the prospect of generating data that can be further used on applications requiring a particular range of datasets. With these capabilities enabled, policymakers and energy planners can use the enhanced data to determine energy access indicators within a program, unlock data forecasting elements, and optimize energy systems when aspects like renewables, low emissions, sizing, and cost are concerned.
