Analysis And Optimization Of High Frequency Of SUV Interior Noise Based On Statistical Energy Method

Automotive interior noise is an important indicator for measuring automotive comfort performance and has been widely concerned by automobile manufacturers and consumers.Automotive interior noise can be divided into low frequency noise,middle and high frequency noise according to the frequency range.At present,Finite Element Method(FEM)and Boundary Element Method(BEM)have been widely used in solving low and medium frequency noise.However,the structures presents a large number of local modes and modal overlaps in the high frequency band,causing the FEM not suitable for solving high frequency noise,thus the statistical energy method SEA has emerged to solve high frequency noise.Based on the basic principle of SEA method,the process of establishing SEA model according to the SUV finite element model is described.The modal density,damping loss factor and coupling loss factor of simple subsystems are calculated by theoretical formula.For complex subsystems,the modal density and damping loss factor are calculated by point admittance real part average method and energy injection method respectively.Sound radiation excitations of engine,excitations of power assembly mount points vibration,Sound Pressure Level(SPL)at driver's right ear and right back row passenger's left ear were tested at the speed of 60km/h,80km/h,100km/h and 120km/h.The pulsating pressure excitations of the vehicle body surfaces at a higher speed were obtained by CFD simulation.model excitations were applied to the SEA model to perform the simulation calculation.the SEA model was validated by comparing the SPL of the driver's head cavity with the experiment results,and the deviation of the simulation was analyzed.Through the analysis of noise contribution,it is concluded that the subsystems which contribute more to the response of sound pressure level of driver's head cavity are front windshield,roof,firewall and front floor.The vibration characteristics of the subsystems such as front windshield in the middle and high frequency under the action of sound load are analyzed.The effects of structural changes on the subsystem itself and the acoustic cavity were discussed by applying enforcing ribs and increasing thickness to the roof subsystem,and found that enforcing ribs cannot improve the SPL of cavity in mid-high frequency.According to the emphasis of sound absorption performance and transmission loss in roof,front floor and firewall sound package,different sound package materials were selected to simulate the sound absorption performance and transmission loss Firstly,then the preliminary selection scheme of sound package was determined.Taking each layer sound packages' thickness of the firewall and front floor as design variables,selecting the SPL at driver's right ear and the total sound packages' weight as optimization targets,Sixty sets of design variable sample were obtained by Latin hypercube sampling,then establish the Kriging approximation model and select multi-objective genetic algorithm optimization method(MOGA),the pareto diagram of sound packages' weight and the SPL of the driver's head cavity was obtained.Selecting the Noraml-Boundary intersection method to find the optimal design scheme of the sound package,Compared with the original sound package,the new sound package weight decreases by 2.18 kg and the SPL of the driver's head cavity decreases by 0.63dB(A),which achieves the goal of reducing the noise inside the car and reducing the weight of the acoustic package.