Abnormal Conditions and their Classification in Photovoltaic Array using Artificial Neural Network

This paper deals with the detection of abnormalities and their classification in photovoltaic (PV) array using artificial neural network (ANN). The PV arrays in the plant have widely varied critical issues including unpredictability and require continuous remote monitoring. Manual module testing is not recommended due to its time-consuming nature, limited accuracy, and potential risks for operators. Therefore, it is essential to develop automated diagnostic tools that can quickly detect faults and classify them to provide effective solutions. In this context, a Python-coded ANN is employed in this paper to detect the abnormal conditions in the PV array using a dataset generated through MATLAB/Simulink platform. The faults considered in this paper are partial shading, short circuit, open circuit and degradation. This approach aims to identify the type and location of faults by training the neural network. By utilising this method, early fault detection can be achieved, enhancing the overall efficiency and reliability of the PV plant. A hardware system has been developed to test the performance of the proposed ANN model.