Multi-sensor Distributed Fault Detection Method Based on Subjective Bayesian Reasoning

Abstract

Each sensor detects information presents problems under uncertainty in computerized numerical control(CNC)machining center fault prediction. Aiming at this problem, multi-sensor distributed fault detection method based on uncertainty reasoning is proposed. The algorithm by using subjective Bayesian reasoning, acquire the local detection device of decision rules, and select the local decision rules suitable to the fusion center, finally a global decision is produced. The complex vertical machining center as an example, the distributed multiple sensor fault detection platform is built. Fault sample is acquired using multi-sensor sample on different machine running state and running environment. Experiments show that in the fault diagnosis system contains a lot of information uncertainty, distributed detection fusion algorithm based on subjective Bayesian inference has the advantages of high recognition rate of fault information, diagnosis speed. Diagnosis error rate of multi-sensor distributed detection fusion algorithm is significantly lower than that of single sensor and the serial structure.