Photovoltaic intelligent fault diagnosis method based on multiple fault characteristic values

The invention provides a photovoltaic intelligent fault diagnosis method based on multiple fault characteristic values, and relates to the related technical field of data processing in a photovoltaic power generation system, and the method comprises the following steps: a photovoltaic inverter scans a photovoltaic module to obtain current-voltage data on an output characteristic I-V (current-voltage) curve of the photovoltaic module; the obtained current-voltage data is preprocessed, the preprocessed current-voltage data is subjected to data processing to obtain fault characteristic values more sensitive to photovoltaic faults, and the fault characteristic values comprise equivalent thermal voltage, I-V curve inflection point factors, maximum power point factors, equivalent series resistance and filling factors; and normalizing the obtained five fault feature values into dimensionless relative unit values, and carrying out fault diagnosis on normalized fault feature vectors. According to the method, fault fea