Bayesian fuzzy inference nets online fault diagnosis of induction motor

A hierarchical Bayesian fuzzy inference nets realtime internal fault diagnostic system for induction motor is proposed. The membership functions and symptom-fault mapping relationship for motor fault diagnosis are obtained from pre-measured site experimental data as well as experts' diagnostic experience/knowledge to distinguish the effect of true fault from various external static factors. With the defined fault fuzzy sets, the hierarchical Bayesian fuzzy inference nets are constructed to carry out the complex motor fault diagnosis procedure. The propagation of probability is used to address the uncertainties involved in detecting and diagnosing stator and rotor faults. The immense difficulties of defining and assigning prior probability, likelihood of sufficiency and likelihood of necessity for each node of the inference nets are effectively solved. Finally the scheme for online diagnosing the fault of multiple motors with improved speed and accuracy is designed and presented. The validity and effectiveness of the proposed approach is witnessed clearly from the testing results obtained. The paper also gives the scope for developing Bayesian fuzzy inference nets and applying the algorithm and methodology for other applications, especially dealing with diagnoses from symptoms to troubles, such as fault diagnosis of IC chips, power systems etc., and even more complicated human disease prognosis.