A machine learning-supported framework for predicting Nigeria's optimal energy storage and emission reduction potentials

Energy sufficiency and the need to reduce carbon emissions have always been at the forefront of global efforts in recent times. This is the motivation of this study which seeks to reduce carbon emissions through the integration of renewable energy sources, by comparing two electricity scenarios for Nigeria by 2050, focusing on the inclusion and exclusion of electricity storage technologies, using a machine learning-supported approach. A Central Composite Design (CCD) was used to generate a design matrix for data collection, with EnergyPLAN software used to create energy system simulations on the CCD data for four outputs: total annual cost, CO₂ emissions, critical excess electricity production (CEEP), and electricity import. Three machine learning (ML) algorithms— multi-layer perceptron (MLP), extreme gradient boosting (XGBoost), and support vector regression (SVR)—were tuned using Bayesian optimization to develop models mapping the inputs to outputs. A genetic algorithm was used for optimization to determine the optimal input capacities that minimize the outputs. Results indicated that incorporating electricity storage technologies (EST) leads to a 37% increase in renewable electricity sources (RES) share, resulting in a 19.14% reduction in CO₂ emissions. EST such as battery energy storage systems (BESS), vehicle-to-grid (V2G) storage, and pumped hydro storage (PHS), allow for the storage of the critical excess electricity that comes with increasing RES share. Integrating EST in Nigeria's 2050 energy landscape is crucial for incorporating more renewable electricity sources into the energy mix – thereby reducing CO₂ emissions – and managing excess electricity production. This study outlines a plan for optimal electricity production to meet Nigeria's 2050 demand, highlighting the need for a balanced approach that combines fossil fuels, renewable energy, nuclear power, and advanced storage solutions to achieve a sustainable and efficient electricity system.