Solar PV Array System Fault Location and Classification using Wavelet and Artificial Neural Network

Abstract

Solar energy is the most abundant, in exhaustible, and environmentally friendly of all renewable energy sources. Because of its advantages, interest in electrical solar PV power generation has grown in recent years. This extensive spread of electrical phenomenon panel manufacture was not accompanied by services such as monitoring, fault detection, and identification to confirm greater gain. This methodology proposes a method for real-time monitoring and fault diagnosis in electrical solar PV systems. This method is based on a comparison of the performance of a defective electrical solar PV module with its right model, which is done by measuring the precise differential residue that is associated with it. The deformations generated on the I-V and PV curves are used to determine the electrical signature of each default. Module to module faults, short circuit faults, open circuit faults, cell to ground faults, and various shading patterns are all taken in to account. The projected method is frequently adapted and applied to a wide range off adults. Back-propagation based Artificial Neural Networks were used to assess the faults situation (ANN). The MATLAB simulation model' s simulation result syndicate satisfactory outcomes for various fault conditions as function of solar irradiation. The entire system is designed and tested in the MATLAB 2015 a software environment for various failure conditions. The system's definition and configuration are based on the basic paper system.