Tobechukwu2000/Electrifying-Nigeria

# Power Sector Multi-objective Optimization in Nigeria using ML In this project, we modeled and compared two scenarios for Nigeria's power sector by 2050: one incorporating grid-scale electricity storage technologies and another without. The objective was to quantify the impact of grid-scale electricity storage on emission reduction and renewable energy integration. Three machine learning (ML) algorithms—support vector regression (SVR), extreme gradient boosting (XGBoost), and multi-layer perceptron (MLP)—were tuned using Bayesian optimization to develop models mapping the inputs (Capacities of generation and storage technologies) to outputs (cost, CO2 emission and critical excess electricity production). A genetic algorithm was employed for multi-objective optimization to determine the optimal input capacities that minimize the outputs.