ABNORMALITY DIAGNOSIS METHOD OF PHOTOVOLTAIC POWER GENERATION, DEVICE, COMPUTER DEVICE AND STORAGE MEDIUM

Disclosed are an abnormality diagnosis method and apparatus for photovoltaic power generation, a computer device and a storage medium. The method comprises acquiring operation data of a photovoltaic station, generating a photovoltaic power generation capacity prediction model by training a preset neural network based on the operation data of the photovoltaic station; determining an abnormal photovoltaic station by performing daily power generation capacity fluctuation evaluation and power generation efficiency evaluation on a plurality of photovoltaic stations using the photovoltaic power generation capacity prediction model; and generating a diagnosis result of the photovoltaic power generation abnormality by comparing the operation data corresponding to the abnormal photovoltaic station with an abnormality condition. According to the method, accurate evaluation of a photovoltaic power generation operation situation and accurate determining of the photovoltaic power generation abnormality are realized, and a brand new perspective is provided for operation management of the photovoltaic station.