Performance Degradation Evaluation Of High Speed Railway Catenary

With the rapid development of high-speed railway,catenary,as an important part of the high speed railway,has become more and more important.Because of the complexity of its structure and the particularity of the environment as a whole,contact line of high speed railway traction power supply system is a weak link in the maintenance.Now the maintenance of catenary is still in the stage of "failure repair",to strengthen the monitoring and detection of the state of catenary and to evaluate the performance of catenary,in accordance with the concept of "prevention mainly,heavy inspection and prudent repair",the "state repair" stage of catenary system is reached,it's a vital task for the development of high speed railways.Traction power supply system of high-speed railway safety monitoring(6C)in the effective use of high speed railway,so that we can accurately detect the characterization of catenary to static and dynamic parameters,the performance degradation assessment of contact line system has strong data support.Through the field investigation and data collection of the high speed rail line from Qingyuan station to Guangzhou the northern station,the field monitoring data of Qingyuan station to Guangzhou the northern station of Wuhan Guangzhou high-speed rail were acquired by 6C system.After the pretreatment was eliminated,the feature was extracted,which laid the foundation for further work.After integrating the acquired eigenvalues as sample data,the improved k-means algorithm is used to divide the state of catenary.The running state of catenary is divided into five different grades.Combined with the actual situation,five kinds of operation conditions are described and evaluated in detail.Then,according to the needs of model establishment,we further divide multiple times to get relevant model parameters.The data is selected as the model training sample to train the hidden Markov model,and a solution is put forward on the possible problems.After determining the parameters of the model,the feasibility and correctness of the model are determined.It is proved that the hidden Markov model can be applied to evaluate the degradation performance of catenary,and the evaluation model of catenary degradation performance based on coupled hidden Markov model(CHMM)is established.Model training samples are trained by contact pressure,pull out value and data under normal condition as model training samples,and input the parameter feature data from other states into the model.In order to verify the correctness of the model,the shape and separability of the output curve in different states are observed.Then,the mean value,variance and synthetic power spectrum value corresponding to contact pressure,height and pull-out value are used to form multidimensional vectors.As input test samples,the performance of catenary is evaluated again.Comparing two models to identify the status of catenary,it is concluded that the evaluation model of catenary degradation performance based on CHMM has higher recognition rate,and its accuracy is higher than that of single chain model.Based on the output of the model,a continuous adaptive alarm threshold is set up to realize the recognition of degradation performance of catenary,and at the same time output its performance index and corresponding alarm level.By analyzing the evaluation results and the actual maintenance condition of the field,the correctness of the results and the feasibility of the evaluation scheme are verified,and the degradation performance evaluation of high-speed railway catenary is realized,which provides a reference for the "state repair" of catenary.