A new fault detection method of PV modules based on BP neural network

Abstract

It is important to ensure the stable operation of PV systems with their increasingly widespread use.For the complex sampling process of most PV fault diagnosis systems, this paper proposes an intelligent detection method of PV cell faults based on the I-V characteristic trend, which takes only the I-V characteristic trend of PV cell as a training sample to achieve high accuracy diagnosis of the three states of normal operation, hot-spot faults and abnormal aging. BP (back propagation) neural network algorithm is used as the training algorithm, and the experimental results show that the average accuracy of the BP neural network with the I-V trend as the training object is 99.3%, which achieves high accuracy fault diagnosis.