Fault diagnosis method and system

The invention belongs to the field of equipment fault diagnosis, and particularly relates to a fault diagnosis method and system. The method comprises the following steps: inputting data corresponding to text description of a fault into a trained fault diagnosis model, and obtaining a feature vector containing semantic information through a natural language processing algorithm module in the model; performing feature optimization compression through a feature optimization module containing an information bottleneck algorithm in the model to obtain an optimization vector output by the feature optimization module; inputting the optimization vector into a classifier module in the model to obtain a fault diagnosis result; the training mode of the model comprises the steps of inputting data corresponding to text description of a fault in a training set into the model, calculating a total loss function, and correspondingly iteratively updating parameters of the feature optimization module until an iteration stop co