Power Quality Enhancement in Hybrid PV-BES System based on ANN-MPPT

Battery energy systems (BESs) assisted photovoltaic (PV) plants are among the popular hybrid power systems in terms of energy efficiency, energy management, uninterrupted power supply, grid-connected and off-grid availability. The primary objective of this study is to enhance the power quality of a grid-tied PV-BES hybrid system by developing an operational strategy based on artificial neural network (ANN) based maximum power point tracking (MPPT) method. A test system comprising a 10-kWh BES and a 12.4 kW PV plant is structured and simulated on the MATLAB/Simulink platform. The hybrid system is validated with three different cases: constant radiation, rapid changing radiation, and real-day solar radiation data from the Turkish State Meteorological Service of Tarsus (Mersin, Turkiye) employing the developed operational strategy. These cases involve the examination of three distinct MPPT methods, analyzing DC-link voltage, battery state of charge (SOC), current, voltage, and system total harmonic distortion (THD). The simulation results indicate that the developed operational strategy with the ANN-MPPT method yields superior THD results in output current and a more stable DC-link voltage. Furthermore, the strategy shows improved convergence speed and reduced oscillations to achieve diverse reference operating points under varying atmospheric conditions compared to conventional MPPT methods. Numerical results demonstrate that the developed operational strategy with the ANN-MPPT consistently maintains THD values below 3% and exhibits a stable DC-link voltage deviation of 1.42% in various charging modes for both rapidly changing radiation and real-day solar radiation data.