Research On Interior Sound Quality Based On The Contribution Of Body Panel

As the car gradually become part of the life,people have become increasingly concerned about the ride comfort.There are many panels in the body structure,and it's an important influencing factor for ride comfort in the car.Panel Acoustic Contribution Analysis(PACA)is an important method in reducing the noise in the passenger compartment.Therefore,based on the panel contribution analysis theory,this thesis was going to research the sound quality of the noise caused by the vibration of the body panel.The main work is as follows:(1)Based on a domestic car,a body structure model was established in CATIA simulation software.The model was imported into HyperMesh,and the finite element model of the body panel and the boundary element model of the interior acoustic cavity were established.The validity of the model was confirmed through experiments.(2)In the vehicle's semi-anechoic room,Idle speed condition and four kinds of uniform speed condition of automobile were simulated on a drum test bench.Head Acoustics software was used to collect the vibration excitation signal at the connection point between the engine mount and the vehicle body,and the noise signal from the driver's right ear.The subjective evaluation of the noise signal samples collected in the experiment was performed by using the grade scoring method,and the irritability value of the sample signals was obtained.In the Head Artemis software,multiple sound quality objective parameters were calculated,and the correlation between the calculated results and the irritability value was analyzed.(3)The prediction models of sound quality based on multiple linear regression theory and neural network theory were established.The prediction model based on BP neural network was selected by comparing the prediction accuracy and modeling difficulty.In this paper,genetic algorithm was used to optimize the initial weight and threshold value of BP neural network,because some of the results of prediction error will be large.Finally,the GA-BP neural network model which can provide higher prediction accuracy was selected as the prediction model of sound quality.(4)The excitation signals collected in the test were inputted into the simulation model,through the ATV calculation of the cavity mode,the contribution of the noise produced by the vibration of each panel to the sound pressure level of the right ear field point of the driver was studied under the key frequency.Combined with the GA-BP neural network model and using simulation methods,the contribution of each panel on irritability was calculated.Through analysis and comparison,we found that due to the coupling between psychoacoustic parameters,there is a difference between the contribution of sound quality and sound pressure level.Panels with a large contribution to sound quality were determined to be ceiling and floor panels as optimized target panels.(5)The target panel noise sound pressure level amplitude was used as a variable,and the ideal irritability value was used as a target.The target noise signal was searched using a genetic algorithm in MATLAB.Through the ATV inverse calculation,the vibration speed of each unit of the target panel was obtained,and the unit with the largest change was obtained by comparing with the original vibration speed.For the unit with large changes,several optimization schemes were designed and the effect of the optimization scheme was tested.The results showed that when the two target panels are optimized simultaneously and the damping thickness is 3mm,the sound quality in the car is better.In order to improve the comfort of the cab,this paper used the finite element and boundary element model as the carrier,and used the panel acoustic contribution analysis method combined with the subjective evaluation method of sound quality to determine the target panel to be optimized in the body design stage.Through the exploration of optimization methods,this paper proposes an innovative optimization scheme.After the panel was improved,the expected optimization goal was achieved.At the same time,it can reduce costs and light weight.