Research on Fault Diagnosis of Photovoltaic Array Based on Virtual Sample Generation Algorithm

To address the issue of limited operational experience and the difficulty in accumulating sufficient fault samples to meet the requirements of fault diagnosis models in photovoltaic array operations, a virtual sample generation algorithm based on Extreme Learning Machine (ELM) is proposed, and a fault diagnosis model is established. This method first utilizes the Local Outlier Factor (LOF) algorithm to identify sparse regions of samples, then uses the ELM to generate virtual samples in these sparse regions, and finally combines virtual samples with real fault samples to train the fault diagnosis model. Simulation verification is conducted using a fault sample set from a photovoltaic power station in Beijing. The results show that the accuracy of the fault diagnosis model reaches up to 95.46%. This algorithm effectively enhances the accuracy of fault diagnosis in photovoltaic arrays.