Font Size: a A A

Research On The Global Out-of-roundness Detection Method Of Urban Rail Wheels

Posted on:2022-08-02Degree:MasterType:Thesis
Country:ChinaCandidate:S T WeiFull Text:PDF
GTID:2512306755952419Subject:Traffic and Transportation Engineering
Abstract/Summary:PDF Full Text Request
With the continuous development and progress of the society,the development of the city has also entered the road of rapid development,in which urban rail transit plays an increasingly important role.However,due to the continuous impact and friction between the wheel and the track of urban rail,the wheel fault is very easy to occur,which seriously affects the safe operation of urban rail train.Therefore,it is of great significance to study the detection method of global out of roundness of urban rail train wheels.On the basis of summarizing the research at home and abroad,this thesis mainly completes the following work.Firstly,based on SIMPACK,the vehicle track coupling dynamics simulation model of urban rail train is established,and the wheel data under different fault conditions are simulated to provide data support for the follow-up content.Secondly,the track vibration signal decomposition algorithm is studied.Aiming at the different noises in the original vibration signal,a noise reduction method based on adaptive morphological filtering is proposed to filter the vibration signal.In order to extract the fault information contained in the vibration signal and deal with the endpoint problem and mode aliasing problem in the decomposition process,an improved local mean decomposition algorithm(LMD)is proposed.The feasibility of the decomposition algorithm is verified by the simulation of vibration signal and actual signal.Thirdly,according to the requirements of the actual project,an automatic detection system for the global roundness of urban rail train wheels is designed,including the overall design of the system,hardware selection,software design,on-site installation,etc.Finally,the wheel fault identification method of urban rail train based on Ada Boost BP neural network is studied.On the basis of improved LMD decomposition,eigenvalues are selected as input vectors to train and learn Ada Boost BP neural network to realize wheel fault recognition.The results of simulation and measurement show that the proposed method can identify the wheel global irregularity of urban rail train.
Keywords/Search Tags:Urban rail train, Wheel global out of round, Vibration signal, Morphological filtering, Local mean decomposition, neural network
PDF Full Text Request
Related items