基于改进BP神经网络的车轮定位参数动态测量

Modified BP ANN Based Dynamic Measurement of Wheel Alignment Parameters

结合人工神经网络(ANN)技术,提出了基于改进的BP神经网络的车轮定位参数动态测量方法,编制了相应的程序,并进行了试验验证。结果表明,通过将车辆前进时的侧滑量作为已训练好改进BP神经网络的输入,根据网络的输出值可以有效地识别出车辆行驶时的车轮外倾角与前束值,从而实现在侧滑试验台上对车轮外倾角和前束值的测定,并依据测定结果有效地指导检修人员进行车轮外倾角与前束值的调整。

Combined with the artificial neural network(ANN)technology,the modified BP neural network based dy-namic measurement method of wheel alignment parameters are brought forward,the corresponding software is pro grammed and test verification is carried out.It is shown that when the side slipping values of the running vehicle are input into the trained BP neural network,cambers and toe-ins of the running vehicle are obtained from its outputs,that is,it is feasible to measure wheel camber and toe-in on side-slipping test rigs.The measured results can be used to guide work ers to effec-tively adjust cambers and toe-ins.