关键词: 车辆工程, 换档品质, 评价方法, BP神经网络, 评价指标

Abstract: The study of the gear shift quality evaluation is required by the development of the vehicle automatic transmission. The evaluation process is unsteady and nonlinear which may be identified and approximated by the backpropagation neural network(BPNN). A BPNN was set and trained by a set of data sample obtained by the definition of the gear shift evaluation indexes. The relation between the subjective and the objective evaluations has been established. The objective description of the gear shift quality evaluation method was performed and simulated using the Matlab/Simulink. A comparison of the simulated results with the test ones shows that the proposed method is in good agreement with the traditional subjective evaluation method, can evaluate the vehicle gear shift quality properly.

Key words: vehicle engineering, gear shift quality, evaluation method, backpropagation neural network(BPNN), evaluation indices