Provided is a photovoltaic assembly pollution diagnosis method based on solar energy generating supervision big data excavation. Through the real-time acquisition of the big data produced by a photovoltaic generating system, the abnormal running parameters of the fault state can be analyzed according to the running parameter trend and rule of the main components under the normal running state, the specific position and fault type of a fault photovoltaic assembly can be deduced, the effective technical support and maintenance strategies can be provided for the generating supervision and management and safe and efficiency running of the photovoltaic power station itself, and the information management level of photovoltaic power plants and group companies can be improved.