An Intelligent Fault Detection Method of a Photovoltaic Module Array Using Wireless Sensor Networks

This study developed a fault diagnosis meter based on a ZigBee wireless sensor network (WSN) for photovoltaic power generation systems. First, the Solar Pro software was used to simulate the 9-series, 2-parallel photovoltaic module array formed with the Sharp NT-R5E3E photovoltaic module as well as record the power generation data of the photovoltaic module array at different levels of solar radiation, module temperature, and fault conditions. The derived data were used to establish the weights of the extension neural network (ENN). The fault diagnosis in the photovoltaic power generation system required extracting the system's power generation data and real-time solar radiation and module temperatures; this study thus developed an acquisition circuit for measuring these characteristic values. This study implemented extension neural network theory using a PIC single chip microcontroller and incorporated the ZigBee wireless sensor network module to construct a portable fault diagnosis meter to assess the photovoltaic power generation system. The experimental results showed that the proposed portable fault diagnosis meter based on the extension neural network for the photovoltaic power generation system possessed a high level of accuracy in fault identification.