Research On Vehicle Sound Quality Evaluation Technology And Method

Sound Quality evaluation technique is one important branch of environmental acoustics. It is closely linked with physiology, psychology, physical acoustics and other disciplines. The emergence of objective evaluation technique adapted the current development needs of sound quality evaluation studies, and provided effective computing means to sound quality. For sound quality study, subjective evaluation is an accurate method which could reflect the sound quality preference directly. Due to the time, effort and other practical problems, the experiment was carried out with greater difficulty and complexity. Although the emergence of objective evaluation technique maked up the lack of subjective evaluation and the results were relatively simple, the calculation accuracy was to be improved. In the research of objective evaluation, scholars at home and abroad were used psychoacoustic parameter values for sound quality analysis and evaluation without considering the impact of head related transfer function, as well as the acousitic signal process of auditory model, the evaluation results were not very satisfactory, and it was difficult to obtain accurate conclusions. The study has been focused on the calculation of objective evaluation parameters, and carried out the research of head related transfer function and the auditory periphery computational model, aimed to find an effective way to calculate the objective parameters accurately and achieve sound quality preferences evaluation. Base on the sound quality objective evaluation technology, the artificial head model and auditory model were analysed and discussed respectively, the artificial head design method base on the head-related transfer function and the construction method of auditory periphery computational model were also proposed. On this basis, an objective evaluation method based on artificial head and auditory periphery computational model was proposed. The application of artificial head model simulated the transmission process from sound field to ear, and the application of computational model simulated the sound signal transmission process of auditory periphery system. The feasibility and superior of the objective evaluation method were proved through the analysis and study of theoretical analysis, physical modeling and the examples of verification. These major research works accomplished are as follows:Firstly, the structure and function of human auditory system were described, and the key features of the auditory system were analysed including the masking effect and binaural effects. Main characteristics and major research methods of sound quality evaluation techniques were described including the subjective evaluation method and objective evaluation method. The psychoacoustic parameters of sound quality evaluation were described importantly. It provided the theoretical groundwork for the following research.Secondly, the main features of sound source location were analysed including the binaural time difference, binaural level difference and head related transfer function. The select method of relevant physiological parameter and its relationship with head related transfer function were concluded. The implement methods of artificial head model based on head related transfer function were proposed. Through the measurement and analysis of head related transfer function to artificial head, the simulation functions of the artificial head were verified. Meanwhile, the auditory periphery computational model was proposed, it realized the simulation of middle ear and inner ear transmission. Combined with artificial head model, the auditory periphery computational model could process the vehicle interior noise; the simulation results validated the accuracy and usefulness of the model.Then, the combined research of artificial head and auditory periphery computational model were carried out. A sound quality objective evaluation based on artificial head and auditory periphery computational model was proposed. The acoustic signal processing processes were elaborated and the psychoacoustic parameter description of objective evaluation was analysed. This part focused on the calculation of binaural loudness calculation process and the weighting method, and as a basis for the model calculation of sound quality objective evaluation parameters. With the same noise signal samples, the objective evaluation method calculation results and the ArtemiS software outputs of the Head Acoustics company were compared,the results showed that the objective evaluation method with superior frequency analysis and computing power, through the mend of artificial head and auditory periphery computational model, the accurate calculation of psychacoustic parameters to noise signal were realized. Contrast caculation examples showed that the error rate of objective evaluation method results and ArtemiS software outputs were remained below 2.3%.Finally,the correctness and accuracy of objective evaluation results which taked the subjective evaluation tests as standard were verified. The vehicle interior noise signal of steady-state conditions were gathered and processed. Based on the subjective evaluation experiments and data analysis, the correlation analysis and calculation between objective evaluation method results and subjective evaluation performance test results were conducted, the effective correlation coefficient calculation results are more than 0.87, it indicated that the linear correlation of the objective and subjective evaluation results was good,and the performance of the objective evaluation method which could be used to the sound quality analysis and calculations were verified. Objective evaluation method was an algorithm computing model; it reduced the cost of the experiments and improved the calculation efficiency. So a good analysis method for sound quality evaluation was provided.The innovation of the research are that through the study of head-related transfer function and auditory characteristics, a sound quality objective evaluation method based on artificial head and auditory periphery computational model is presented. The significance of this method are that the impact of head related transfer function was increased compared to the previous method, and some parameters of auditory periphery computational model were improved to make it more suitable for the analysis of interior noise signal characteristics. The method has some practical value in sound quality evaluation technology.