摘要
在MATLAB软件中,通过选取汽车长度、宽度、高度、整备质量、排量、被动安全装置数与价格作为输入,相对应车型的C-NCAP试验评价值作为输出,建立BP神经网络来对汽车被动安全进行评价。经过50组车型数据的训练,网络模拟误差在0.1%范围内;利用22组车型数据对该评价模型进行验证,预测结果与C-NCAP官网相对应车型碰撞试验评价值的误差控制在0.17%范围内。利用GUI工具基于BP神经网络预测模型创建汽车被动安全评价软件,并对A,B,C三辆汽车进行被动安全评价,其值与实际碰撞试验值接近并能正确区分各车被动安全的好坏。
An automotive passive safety evaluation system based on MATLAB was set up,by use of BP neural network,to select the car＇s length,width,height,curb weight,displacement,number of passive safety devices and price as input,the corresponding C-NCAP test value as output,after using 50 sets of training data to train the model,the network simulation error was in the range of0.1%.Then 22 sets of data were used to test the system,compared with the C-NCAP test values,the actual prediction error was in the range of 0.17%.The GUI tool was used to create automotive passive safety evaluation software,and the passive safety of three vehicles A,B and C were tested,the tested values were accordance with the results from the actual crash test.
出处
《机械工程与自动化》
2015年第1期61-62,共2页
Mechanical Engineering & Automation
