Predictive battery dispatch for 24-hour utility-scale solar power integration at Kabulasoke Solar Power Plant in Uganda

Abstract

The growing reliance on renewable energy in Uganda demands innovations that ensure continuous and stable electricity supply, especially in the face of variability in solar power generation. While grid-connected solar power plants like the 20 MW Kabulasoke Solar Plant contribute significantly to Uganda’s energy mix, they currently lack battery energy storage and intelligent control mechanisms to support night-time and poor-weather operations. This limitation reduces the reliability of solar energy as a 24-hour power source. This project proposes a predictive battery dispatch framework for utility-scale solar power systems, enabling continuous dispatch of energy to the grid. The system forecasts short-term solar generation and dynamically manages battery charge and discharge based on predicted generation, load demand, and state of charge (SOC). The predictive model addresses the limitations of rule-based control by proactively storing energy during surplus conditions and supplying it during night or low irradiance periods. A MATLAB/Simulink simulation was developed to evaluate the system's performance over a 24-hour cycle. Results demonstrate that predictive dispatch improves energy reliability, maintains battery SOC within operational limits, and ensures uninterrupted night-time load support. Economic analysis further reveals that while battery integration adds capital cost, it significantly reduces long-term curtailment and backup costs. The study highlights the practical value of intelligent battery dispatch in Uganda’s solar infrastructure and presents a scalable solution that enhances grid stability, energy security, and sustainability.