Maximum Power Point Tracking using artificial intelligence

The photovoltaic systems are increasingly popular as green energy due to energy crisis and environmental issue. The PV arrays are still costly, and to make them cost-effective they are required to operate with maximum efficiency irrespective of temperature, solar irradiance and load condition. The conventional Maximum Power Point Tracking (MPPT) acts as an interface between solar PV and the load, and helps to extract maximum power by matching the impedance of the load to that of PV. However, the conventional MPPTs lack robustness over weather conditions and require some tracking time. In this work, a neural network based method is proposed which doesn't require any tracking. It can deliver the right signal to dc-dc converter to reach the MPP without any delay or iteration. Thus, the proposed system is much faster than the conventional ones. Our simulation results corroborate the robustness of the system over wide range of temperature, irradiance and load conditions.