Prediction Research And Simulation Optimization Of A Certain Car Door Closing Sound Quality

For cars,the door closing sound has not only the characteristics of noise,but also the characteristics of quality.To control the closing sound of a car,it is necessary to strictly control the size of its sound pressure level and improve its sound quality.The method of evaluating the quality of door closing sound is usually subjective and objective.However,most evaluation and optimization analysis of door closing sound quality can only be carried out after the actual vehicle is manufactured,which is poor in flexibility and timeliness.Considering the current car door closing sound quality evaluation and optimization problems,this paper studies the sound quality evaluation of car door closing sound,the model establishment of door closing sound quality prediction,the collision simulation analysis and optimization analysis of door closing sound quality,etc.The main research contents of the article are as follows:(1)Noise collection equipment was used to collect the door closing sound signals of 18 different car models.Establish a sound sample library of sound quality when the door is closed and perform subjective evaluation of the door-closing sound signal to obtain a subjective evaluation score.Then,the objective evaluation parameters of the psychoacoustics in the door closing sound signal,such as loudness,sharpness,roughness and jitter,are extracted,and the quality of the door closing sound is subjectively and objectively evaluated.(2)Taking loudness and sharpness as input and subjective evaluation score as output,BP neural network prediction model and support vector machine prediction model are obtained.A comparative analysis of the prediction accuracy of these two prediction models shows that the prediction results of the established support vector machine prediction model are more accurate.The establishment of the support vector machine prediction model lays the foundation for the sound signal prediction obtained by the subsequent door closing sound simulation.(3)In order to solve the problem that most of the research on the quality of door closing sound cannot be separated from the actual vehicle,this paper applies the simulation methods commonly used in the analysis of car safety collisions in the study of car door closing collisions.Established the finite element model and boundary element model of the car door.After that,the accuracy of the simulation model was verified by the door-closing vibration and noise experiment.Because the door structure has a great influence on the door closing sound of a car,this paper takes the door inner panel as a design variable,and takes the door modal frequency as the design goal to optimize the door inner panel structure.(4)Establish a boundary element simulation model and calculate the sound pressure value and sound quality objective evaluation parameters obtained before and after the car door optimization.Then input the obtained objective evaluation parameters into the support vector machine prediction model to obtain the quality scores of door closing sound before and after optimization.The results show that the optimization of the car door inner panel structure can reduce the loudness and sharpness and improve the sound quality prediction score.The quality of car closing sound has been improved.The method described in this article can provide reference and guidance for the quality analysis and optimization of car door closing sound,and has certain value and significance.