A study on the evaluation method of vehicle door closing sound quality based on artificial neural network model

作者：Zhang, Bo; Bi, Chuanxing; Xu, Liang

来源：Qiche Gongcheng/automotive Engineering, 2011, 33(11): 1003-1006.

• Artificial neural network modeling
• BP artificial neural network
• Objective evaluation
• Sound Quality
• Subjective and objective evaluations
• Subjective evaluations
• Tonality
• Vehicle doors

The evaluation method of sound quality is studied based on BP artificial neural network and using the objective evaluation data of vehicle door closing sound. The BP artificial neural network is trained based on the subjective and objective evaluation results of the sound samples collected, and then the subjective evaluation prediction is conducted on the trained network. In addition, the objective evaluation parameters of door closing sound as the inputs of artificial neural network model are also studied. The results show that the method is effective, and tonality is an important characteristic parameter of vehicle door closing sound. Taking tonality as one of the input parameters not only can improve network convergence rate, but can also reduce network prediction error.

• 出版日期2011
• 单位合肥工业大学