Quantitative Assessment Technology Of Locomotive Bogie Maintenance Quality Based On Sound

Locomotive bogie,as an important component of locomotive,needs to be decomposed and overhauled during locomotive maintenance.In order to check the maintenance quality of locomotive bogie gear box and the reliability of bearing,running-in test should be carried out on the running and closing test bench,so as to simulate the actual running situation of the bogie.This paper presents a quantitative evaluation method of bogie maintenance quality based on sound signal.The noise is reduced and multi-eigenvalues are extracted.The abnormal detection of bogie is realized by using the Gauss mixture model.The fault diagnosis of bogie is carried out by using the spectral kurtosis method of the average frequency band kurtosis,and the evaluation indexes of the reliability of bogie bearing and the maintenance quality of gear box are put forward.This subject takes the maintenance quality of locomotive bogie gear box and bearing reliability as the research object,and combines the monitoring and evaluation of the current situation and demand of bogie running-in test.The main research contents include:(1)Understand the shortcomings of quantitative evaluation methods for locomotive bogie maintenance quality at this stage.Research the common failure forms,failure characteristics,sound signal generation mechanism and its relation with vibration signal of bogie bearing and gear box are studied.(2)The quantitative evaluation method of locomotive bogie maintenance quality based on sound signal is studied.Research on signal noise reduction,feature extraction,anomaly detection,fault diagnosis and maintenance quality grade evaluation.The bogie acoustic signal characteristics,noise reduction methods and multi-feature extraction are studied;the bogie anomaly detection method based on Gauss mixture model is studied;the bogie fault diagnosis method is studied;and the evaluation index of the reliability of bogie bearings and the quality of gear box maintenance is also studied.According to the research results,a quantitative evaluation method based on the combination of wavelet packet analysis,Gauss mixture model and spectral kurtosis of average frequency band kurtosis is proposed,which is called the quantitative evaluation method of bogie maintenance quality based on sound signal.The bearing fault simulation test rig is used to carry out operation tests of different fault degrees,which verifies the feasibility of the method.(3)The monitoring and maintenance quality grade evaluation system of locomotive bogie running-in test is built,and the signal processing method proposed in this paper is used to realize the real-time monitoring,fault diagnosis and maintenance quality grade evaluation of locomotive bogie running-in test.