Anomaly detection method for photovoltaic equipment

The invention provides a photovoltaic equipment anomaly detection method, and relates to the field of photovoltaic technology, and the method comprises the steps: obtaining preprocessed photovoltaic sample data; constructing an anomaly detection model, wherein the anomaly detection model comprises an anomaly detection module, an anomaly classification module and an anomaly recognition module; inputting the photovoltaic sample data into an anomaly detection module to obtain abnormal sample data; determining the exception type of the exception sample data, and judging the exception sample data based on the exception type; identifying the abnormal sample data based on an abnormal identification module to obtain an influence range of the abnormal sample data; iteratively training the anomaly detection model until an iteration stop condition is reached, and obtaining a trained anomaly detection model; and obtaining new photovoltaic data, and performing anomaly detection on the new photovoltaic data by using the tr