Research On Motor Fault Diagnosis Using Multi-Source Information Fusion Technology

With the development of modern science and technology,electric machines play a more and more important role in modern industrial plants.The risk of machine failing can be remarkably reduced if normal service conditions can be arranged in advance.Uncertain problems are key problems in fault diagnosis fields,which are resulted by many important reasons,including complex diagnosis objects,limit test means,and the inexact diagnosis knowledge,etc.Motor is large electromechanical equipment,which has complex, correlative relation exiting in its units,and the units are full of uncertain factors and information.So the fault of motor is possible multi-kinds,correlative.In the all,the most important problem of solving motor fault diagnosis is to solve the uncertain problem. There are many experts researching on the methods,and has obtained that,Bayesian network based on Bayesian theory is the best method for solving uncertain problems now.This dissertation presents an idea to handle uncertainties in induction motor fault diagnosis with multi-source information fusion technology.Based on detail synthesis of existing results,as uncertainty theories are fundamental in information fusion technology, this dissertation has a detail review of information fusion related uncertainty theories. Various uncertainties exist in induction motor diagnosis methods,the solutions always rely on validity and precision of sensors,accuracy of the signal processing methods.Traditional methods have an inherent shortcoming because they mostly base on single sensor information.Due to existing the uncertain factors,the accuracy of the fault diagnosis is ensured difficultly.However,Bayesian network fusion method can improve greatly the accuracy of the motor fault diagnosis.For solving uncertain problems,the fault diagnosis structure model and function fusion model was proposed,the models were based on the information fusion technology and the research of uncertain problems in fault diagnosis process,and the diagnosis object was motor.Bayesian network fusion algorithm was proposed for the knowledge expression and constitution of the models,which were researched deeply.To analyze the limitation of traditional method of fault tree and the difficulty to construct traditional Bayesian networks, a new method which combines fault tree and Bayesian networks is proposed to construct the model of failure diagnosis,namely Diagnostic Bayesian Networks.It also expatiate the philosophy and algorithm to fault diagnosis strategy optimization method of fault tree and Bayesian networks.Because the Bayesian network was composed of nodes and arcs,a diagram search algorithm was adopted for search the fault nodes.Through confirming,the complexity degree of the reasoning algorithm of this paper was multinomial,and the uncertain problems of fault diagnosis were solved by availably and exactly.