Research On The Intelligent Perception Method Of Catenary Based On The Vibration Response Of The Pantograph

With high efficiency,energy conservation and comfortable security,rail transit has become one of the most important transportation modes in China.As running speed and traffic density has increased,train vehicle dynamic interaction between catenary and pantograph has increased,so the catenary-pantograph system of rail transit vehicles is proposed requirements is higher and higher also.As the main power source of modern rail vehicle,the contact catenary is directly affected by the failure.Open-air set up and huge impact load to a large extent affects the pantograph catenary system of steady flow,thus ensuring the stability of pantograph catenary system,timely grasp the meaning of catenary of state is more and more important.This paper is to model the pantograph-catenary system,analyze the influence of dynamic parameters on the vibration of the pantograph-catenary system,and study the relationship between the vibration response of the electric pantograph and the state of the contact catenary,and provide a new way for the diagnosis of the contact catenary fault diagnosis and daily maintenance work.This paper makes an in-depth analysis of the coupling of the pantograph-catenary by establishing the contact catenary-the electric pantograph model,and considers the important role of the contact catenary components such as hanging string,locator and wire clip in the pantograph-catenary model,and studies the vibration characteristics of the pantograph-catenary system under different dynamic parameters.Several kinds of dynamics simulation analysis of the contact model of main parameters,such as mass,stiffness and damping,hanging string locator stiffness,damping,contact wire cross section parameters of pantograph catenary vibration dynamics such as response,at the same time for the state of the catenary intellisense provides the simulation data.Through the establishment of the pantograph catenary vertical coupling vibration model,the dynamic simulation of the failure of the electric pantograph and the contact catenary,such as the loose and the contact catenary of the electric pantograph frame,is carried out.The electrodynamic response characteristics of different fault types are analyzed.Support vector machine(SVM)algorithm based on different fault types of the pantograph response of the corresponding classification,by extracting the former research in this paper the dynamic response of pantograph catenary troubles as the training sample,characteristics by support vector machine(SVM)classification,automatically by the machine learning to identify response and classification.On this basis,respectively using particle swarm algorithm and gecatenaryic algorithm for support vector machine(SVM)parameters optimization,comparing two optimization algorithm for support vector machine(SVM)algorithm optimization results,so as to improve the accuracy of catenary fault classification.This catenary fault identification method based on support vector machine(SVM),implements the contact line of the structure of the two types of fault diagnosis,the proposed new method was applied to simulation of catenary fault recognition,pantograph dynamic response analysis of the corresponding fault features,fault characteristics into the optimized support vector machine(SVM)classification algorithm.The results show that the accuracy of fault identification is 83.3% and there is no leak detection,indicating the effectiveness and practicability of this method.The paper has realized the state of the contact catenary based on the dynamic response of the electric pantograph,and provided a new way for the detection of electric locomotive contact catenary.