Analysis And Control Method For Transient Vibration And Noise Discretized Transfer Path Of Vehicle Door Slamming Event

With the development of the domestic automobile industry and the continuous improvement of people’s living standards,consumers have increasingly stringent requirements for vehicle comfort.Noise,vibration and harshness(NVH),as an important part of comfort,are increasingly valued by consumers.In recent years,the traditional vibration and noise control can not meet the strict requirements of the market even though it meets the mandatory regulatory requirements.As an important cover of the vehicle,the door allows the vehicle body to form a sealed seating space.The sound of vehicle door slamming event not only has the characteristics of noise,but also has the characteristics of sound quality.The sound quality of vehicle door slamming event is one of the important indicators to measure the quality of the vehicle.Therefore,consumers will always open and close the door many times when purchasing a vehicle to subjectively feel the sound quality,so as to decide whether to buy this model.The sound quality of the vehicle door slamming event determines the consumer’s willingness to purchase a vehicle.Taking the vehicle door slamming event as an example,this paper studies the analysis and control method of the vehicle door slamming transient vibration and noise transfer path,and carries out the following research work:(1)A-weighted sound pressure level,loudness,roughness,sharpness,and sound pressure level are quantified according to the characteristics of transient noise with strong nonlinear coupling,wide frequency range,and short duration time in the driver’s ear noise under door slamming conditions.Based on the analysis of the door slamming noise,the applicability of the main impact time and the low frequency duration time is proposed.The subjective evaluation test of sound quality is carried out,the door slamming noise samples are scored by the graded scoring method,and the subjective evaluation samples with poor consistency are eliminated by the Spearman correlation coefficient method,and finally the irritability value of each noise sample is obtained.(Genetic Algorithm-Back Propagation Neural Network,GA-BPNN)is used to establish the nonlinear mapping relationship between the objective sound quality evaluation index and the subjective evaluation results.By comparing and verifying the prediction error of the subjective sound quality of the noise samples,the accuracy of the constructed neural network is verified,which lays a foundation for the subsequent prediction of the sound quality of the transfer path.(2)Aiming at the problem of abnormal vibration at the door slamming event when the left front door glass is in the bottom position,a discretized time-domain transfer path analysis(DTTPA)model of the door slamming event is established.A vehicle test and a bench test are designed to collect vibration data and transfer function characteristics under door slamming conditions,respectively.The signal is processed by slicing,zero-filling,fast Fourier transform(FFT),inverse matrix solution,Inverse Fast Fourier Transform(IFFT),overlapping,adding to obtain the door slamming transient impact load,and the feasibility of the method is verified by the transfer path contribution synthesis method.The time-frequency characteristics of door slamming transient impact load are analyzed.By comparing the consistency between the transfer function characteristics and the contribution of the transfer path,it is found that the order of the contribution of the main transfer path is consistent with the order of the transfer function characteristics,but has nothing to do with the order of the magnitude of the transient impact load.It shows that the vibration magnitude can be suppressed by improving the structure and reducing the transfer function curve,which has been verified by experiments.(3)A numerical and experimental analysis framework of the vibro-acoustic coupling model is proposed for the door closing impact event.The framework consists of DTTPA,semi-constrained modal analysis based on finite element analysis and experimental testing,and boundary element analysis.A semi-constrained modal test of the vehicle door is designed.The transient impact load of the secondary transfer path is solved by combining the test and simulation,and the feasibility of the combined method is verified by the contribution synthesis.The accuracy of the test and simulation of the semi-constrained mode of the vehicle door is verified by modal confidence criterion analysis.The vehicle test is used to verify the correctness of the transient vibration response of the door slamming event obtained by the modal superposition method.The sound pressure amplitude at the microphone is obtained by using the indirect boundary element method.By dividing the vehicle door system into the door panel system and the guide rail system,the vibro-acoustic coupling value of the door slamming event is obtained.The results show that the sound pressure level generated by the door panel system and the guide rail system is greater than that of the entire vehicle door system,which is caused by the vibro-acoustic suppression phenomenon between the door panel system and the guide rail system.(4)A multi-objective optimization process of vehicle door slamming event based on dynamic performance index is proposed.According to the unique characteristics of vehicle door slamming event,the input point inertance test is designed.The mass,natural frequency and maximum amplitude at the target point of vehicle door are taken as the objective function,the connecting point dynamic stiffness of the guide rail is taken as the constraint function,and the main components of the door are taken as the design variables.Among them,the objective function and the constraint function have passed the test and simulation verification,and both meet the engineering application error.The time-consuming finite element iteration method is replaced by the response surface method,and the accuracy of the surrogate model is verified by four error analysis methods.The three-dimensional Pareto frontier solution of the door slamming event is obtained by using the multi-objective particle swarm optimization.The results show that the dynamic indexes obtained by the multi-objective optimization are better than the original design.(5)Aiming at the door closing impact event,the noise analysis and optimization process with mixed uncertainty is proposed.The uncertainty of door structure and external factors are considered at the same time,and the mixed uncertainty models of probability uncertainty and interval uncertainty are introduced respectively.The whole vehicle model of interior noise analysis is established,the noise transfer function from the excitation point to the driver’s ear is obtained,and the contribution of the transfer path from the excitation point to the driver’s ear is obtained.The feasibility of the proposed method is verified using the contribution synthesis response.Using the above established transient sound quality prediction model of door slamming event,an uncertainty analysis method of refined noise transfer path is proposed.The results show that compared with the deterministic optimization results,the reliability optimization results slightly sacrifice the target performance,but significantly improve the reliability level.