Implementation of a Fault Diagnosis System Using Neural Networks for Solar Panel

In this paper, we propose a fault diagnosis system for the solar panels of solar-powered street lights that uses an adaptive resonance theory 2 neural network (ART2 NN) and a multilayer neural network (MNN). To diagnose a fault in a solar panel, we use the open-circuit voltage with respect to the duty cycle as input for the two neural networks. As a result, we can use them to double check the fault diagnosis for the solar panel. In addition, we present a graphical user interface for the proposed solar panel fault diagnosis system. The fault diagnosis system we propose has the potential for application in similar systems and devices.