Research On The Models Of Fault Prediction And Health Status Assessment For Switch Machine

As one of the basic equipments widely used in railway signal system in China,S700 K electric switch machine plays an important role in ensuring traffic safety and improving transportation efficiency.For a long time,the "fault repair" and "regular repair" used by railway power department can hardly meet the development needs of high-speed,heavy load and high density of railway line.On account of the switch machine is in the outdoor complex environment all the year round,the occurrence of mechanical fault has obvious characteristics of fuzziness and randomness.With the development of Intelligent Transportation System(ITS),this paper studies switch machine fault prediction and health status assessmant by introducing the Fault Prognostics & Health Management(PHM)theory.Firstly,the system analysises S700 K switch machine structure,function and system action process.The changes of fault characteristics,modes and characteristic parameters are obtained in detail,which laids a foundation for the analysis of PHM theory.On the one hand,in view of the switch machine structure is complicated and the working environment obtains relatively bad characteristics,keeping high mechanical failure rate characteristics,this thesis focuses mainly on mechanical failure research.The Hidden Semi-Markov Model(HSMM)with a good description of its whole life cycle states is selected for process modeling.In view of the disadvantages of HSMM model in the actual parameter estimation process,which is easy to be trapped into local optimization.This thesis chooses Improved Particle Swarm Optimization(PSO)algorithm to realize the optimization of HSMM model in order to improve the prediction accuracy of the model.D-PSO algorithm and SA-CPSO algorithm are introduced respectively.Through the fitness test,the HSMM optimized by SA-CPSO PHM model of switch machine is finally established to evaluate the equipment state and predict the remaining useful life(RUL).On the other hand,in view of the fuzziness and randomness of the faults of the switch machine,it is difficult to conduct accurate system modeling.In this thesis,the degradation model of switching machine is established based on the whole life cycle health status degradation process.At the same time,the fuzzy theory is introduced to establish the subjective and objective comprehensive evaluation system of switch machine health status assessment by combining 9-scale Analytic Hierarchy Process(AHP)and entropy weight method.Finally,the validity of the proposed method is verified by model simulation.This thesis aims to provide new idea for intelligent operation and maintenance of railway signal system.