bearing fault intelligent diagnosis method for digital twin system

The invention discloses a bearing fault intelligent diagnosis method for a digital twin system, and belongs to the technical field of bearing fault unbalance detection. According to the method, the bearing fault diagnosis effect is improved under the condition that normal data and abnormal data in a digital twin system are unbalanced under the actual condition and original data are not expanded. The method comprises the steps of monitoring a bearing vibration signal of a target bearing in real time by a digital twin system; inputting the current bearing vibration signal of the target bearing into the bearing fault diagnosis network, and obtaining the current fault detection result of the target bearing based on the output of the bearing fault diagnosis network. The bearing fault diagnosis method is used for diagnosing the data imbalance phenomenon in the actual scene of the bearing, the diagnosis effect on the fault data can be improved in the actual scene with normal and fault data imbalance, and the health