Fault Prediagnosis, Type Identification, and Degree Diagnosis Method of the Photovoltaic Array Based on the Current-Voltage Conversion

The study of fault diagnosis technology is significant for the operation and maintenance of photovoltaic (PV) power plants. A prediagnosis of PV array faults based on current-voltage (I-V) conversion, type identification, and degree diagnosis method is proposed. I-V conversion method is used to eliminate the influence of ambient conditions and provide a reliable data source for fault diagnosis. First, fault prediagnosis is accomplished by comparatively analyzing the difference between reference and measured I-V curves. When the PV array is faulted, fault type is then identified by the unique characteristics of common four types of faults. Finally, the features of the I-V curves are extracted. The extracted features and the corresponding fault labels are used to train the classification model to realize the fault degree diagnosis. The experimental section provides the performance evaluation method of the feature extraction algorithm. The fault prediagnosis results can provide the reliable evidence for the subsequent fault diagnosis. The experimental results based on simulated and measured I-V curves show that the accuracy of both fault type identification and fault degree diagnosis is above 99%. The effectiveness and practicality of the proposed methods are adequately verified.