Modeling and Control of Photovoltaic Array and Wind Turbine for Hybrid Power Generation Using Artificial Neural Network

This thesis work concentrates on adaptive control of power transmission from the renewable energy resources such as photovoltaic (PV) system, and wind turbine (WT) system to the grid with the help of artificial neural network (ANN). In this study, a maximum power point (MPP) tracking algorithm from the photovoltaic system and the wind turbine system utilizing the feed-forward artificial neural network were evaluated. Real-time values of solar irradiance and atmospheric temperature were given as inputs to design the neural network algorithm used in the tracking the maximum power of the photovoltaic system. Variation in wind speed was considered to formulate the extraction of the maximum power of the wind turbine. The lithium-ion battery was used as a backup power system. A controlled Battery Energy Storage System (BESS) and bi-directional power flow from the renewable energy resources to the grid were studied. Also, a three phase inverter control based on the Space Vector Pulse Width Modulation (SVPWM) technique was designed to convert the DC power to the three-phase alternating power and was transmitted to the grid. Lastly, the maximum power from the photovoltaic system, wind turbine system and the state of charge (SOC) of the battery were used as inputs to the neural network for power tracking of the overall hybrid power system with total harmonic distortion analysis. MATLAB/Simulink software was used to validate the proposed system