Intelligent equipment fault diagnosis method based on data and mechanism hybrid drive

The invention belongs to the technical field of federated learning, and discloses an intelligent equipment fault diagnosis method based on data and mechanism hybrid drive, which comprises the following steps: step 1, each source client performs initialization operation on a local model; 2, training a local model by using source domain data; 3, training the local model by using the transition data; 4, the central server updates parameters of the feature extractor and the classifier of the global model according to the local model; 5, performing reverse verification on the source client task by using the global model, and updating the local model in the source client by using the task verification loss; and step 6, circulating the steps 2-5 to reach a set number of times. The invention provides the intelligent equipment fault diagnosis method based on data and mechanism hybrid driving, which can reduce covariant drift caused by domain difference and correct a network concerned area, and improves the accuracy of