基于神经网络的DCT换挡品质评价系统的开发  

Design of Evaluation System for DCT Shift Quality Based on BP Neural Network

作　　者：张晓明[1] 吴光强[1] 张德明[1] 

机构地区：[1]同济大学汽车学院,上海201804

出　　处：《汽车科技》2008年第6期38-41,共4页Auto Sci-Tech

基　　金：上海市科委重大攻关项目(06dz11002)

摘　　要：通过对DCT换挡过程的分析,确定了换挡时间、车辆加速度、换挡冲击度和发动机转速波动量等4个物理量作为客观评价指标。依据BP人工神经网络的基本原理,应用Matlab软件的神经网络工具箱函数,建立了用于评价车辆换挡平顺性的神经网络模型。利用GUI工具箱建立人机交互界面。通过内置的三种评价标准以及可以自行设计评价标准的方法,大大增强了该评价系统的灵活性和针对性。最后,对该评价系统进行了总结。Shift time, vehicle acceleration, shift jerk and the change of engine rotation speed are determined as four objective evaluation indices through analyzing the shift process of DCT. the ANN model used for evaluating shift smooth aspect is build by using the ANN toolbox in MATLAB software based on the principle of ANN. the interface is created using GUI box in MATLAB environment. The flexibility of this system is enhanced because of the internal three evaluation standard and expand standard created by the customer. At last, a summary of the evaluation system was made.

关 键 词：人工神经网络 DCT 换挡品质 评价系统 

分 类 号：TP391.76[自动化与计算机技术—计算机应用技术]