Distributed photovoltaic power generation abnormity diagnosis method and system

The invention relates to a distributed photovoltaic power generation abnormity diagnosis method and system. The method comprises the following steps: acquiring distributed photovoltaic data; selecting features according to the acquired distributed photovoltaic data; performing data preprocessing; setting complexity and model parameters of an XGBoost set model, and adding a tree in the model to enable an objective function of the model to be optimal so as to construct a prediction model; according to the selected features, training and testing the prediction model, iterating the model, adjusting various parameters of the model, and obtaining the prediction model after parameter adjustment; inputting photovoltaic power generation data to be measured into the prediction model after parameter adjustment, and performing abnormality diagnosis on distributed photovoltaic power generation; the system and the method are based on the same concept and comprise five modules of data selection, feature selection, model con