Bearing Fault Diagnosis Method Based on Fuzzy Petri Nets with Adam Optimization

To address the problem that three-phase asynchro-nous motor is difficult to maintenance support due to frequent faults and strong fuzziness of faults. A fault diagnosis method using fuzzy Petri nets (FPN) based on adaptive moment estimation (Adam) optimization is proposed. First, using fault tree analysis method, a bearing diagnosis model based on Adam-FPN is constructed. Then, adopting transition factors instead of traditional expert experience, the Adam algorithm and the Bp algorithm are combined into the Petri net to optimize the inference algorithm and achieve accurate prediction of bearing faults. Finally, the bearing fault model is used as an example for forward and reverse inference analysis and compared with other methods to verify its effectiveness and superiority.