Font Size: a A A

Study On Vehicle Starting Motor Noise Characteristic And Noise Source Identification

Posted on:2018-11-22Degree:MasterType:Thesis
Country:ChinaCandidate:C Q GongFull Text:PDF
GTID:2322330515469015Subject:Vehicle Engineering
Abstract/Summary:PDF Full Text Request
The starting motor of the automobile is close to the cab,and it will emit loud noise when it stars which will affect the comfort of the vehicle.The noise source identification technology can effectively identify the noise source thus people can take targeted measures to reduce noise.In this paper,three kinds of vehicle starting motors were treated as the research object,aiming at noise problem of starting motor to anlyze its noise characteristics,and the EEMD-FastICA-STFT method are proposed to study the niose source indentification of motor noise,the main work of this paper includes:Firstly,the noise and vibration experiement in constant speed condition and vehicle condition were taking on A and B starting motor to acquire the noise and vibration data.The noise test in variable speed condition was taking on C motor to acquire noise data.Secondly,the noise characteristic of A?B motor were ananyzed,combing the method of vibration and noise analysis,the noise soure of A?B motors were identified.The ensemble empirical mode decomposition(EEMD),independent component analysis(ICA)and short time fourier transform(STFT)were proposed to study the noise identification of starting motor.Firstly,the empirical mode decomposition was used to decompose the single channel noise of starting motor into several Intrinsic Mode Function,and FastICA algorithm was used to extract independent components,At last,short time fourier transform was used for its better time frequency characteristic to analyze the time frequency characteristic of FastICA results,Combining the results and prior knowledge of motor noise,the corresponding relationship between independent components and the different noise sources of motor were determined.Finally,on the basis of EEMD-FastICA method,the noise source indentification system was developed which is special for motor under the Matlab software platform.The main funcitons are as follows:time-domain analysis,frequency analysis,time-frequency analysis,independent component analysis.The humanized design is convenient for operation and can easy to analyze the motor noise source.In summary,the research work of this paper not only provides the basis for the vibration and noise reduction of the starting motor.but also provides a reference for the on-line detection of the starting motor noise.
Keywords/Search Tags:noise source identification, time frenquency analyze, independent component analyse, starting motor
PDF Full Text Request
Related items