This paper presents the development and substantiation of an intelligent control approach for a corona discharge - based water disinfection system powered by renewable energy sources. The aim of the study is to improve disinfection efficiency and energy performance of the system under conditions of unstable power supply by optimizing the operating regimes of the corona discharge. The research combines theoretical analysis, mathematical modeling, machine learning and reinforcement learning methods, and laboratory-scale experiments. The results show that stable corona discharge operation in the voltage range of 5 - 30 kV ensures the generation of ozone at concentrations of 10-40 mg/L and hydroxyl radicals in the range of 10-9 10-6 mol/L, providing microorganism inactivation efficiencies of 90 99.99%. The application of the intelligent control algorithm enables discharge instability to be maintained within ±10-15% under solar and wind energy fluctuations of ±20-50%, while reducing overall energy consumption by 15-35%. The obtained results demonstrate that the proposed approach has significant scientific and practical potential for enhancing the stability and energy efficiency of renewable energy-powered water disinfection systems.