Heating ventilation air conditioning system fault diagnosis method and system

The invention discloses a heating ventilation air-conditioning system fault diagnosis method and system, and the method comprises the steps: obtaining the normal working condition data of a heating ventilation air-conditioning system, and constructing a three-dimensional modeling data set; expanding the three-dimensional modeling data set into a two-dimensional modeling matrix along a batch direction and a process variable direction; dynamic variables are extracted from the two-dimensional modeling matrix, and a co-integration analysis model is constructed according to linear combination of the dynamic variables; a co-integration analysis model is adopted to extract latent variables causing dynamic time varying of the heating ventilation air-conditioning system from the to-be-identified fault data and the historical fault data, and a graph neural network integrated with a self-attention mechanism is trained according to the latent variables of the historical fault data; and classifying the latent variables of