Fault diagnosis of photovoltaic array based on SE-ResNet

Abstract

In recent decades, fault diagnosis of photovoltaic (PV) arrays has become more and more important for the power generation of PV systems. Many traditional artificial intelligence methods have been successfully applied to use fault data samples to build fault diagnosis models, but most rely on manual feature extraction or expert knowledge to build diagnosis models, which is inefficient and may ignore some potentially useful features. Therefore, this paper proposes an SE-ResNet PV array fault diagnosis algorithm based on Residual Network (ResNet) and Squeeze-and-Excitation Network (SENet), and uses Bayesian Optimization (BO) optimize the parameters. In order to verify the effectiveness of the proposed fault diagnosis algorithm, a comprehensive fault experiment was carried out on the PV platform. Experimental results show that the model has achieved high overall performance in terms of accuracy, generalization performance, reliability and training efficiency, and has high practicability.