基于混合田口遗传算法的磁流变半主动悬架模糊控制

Fuzzy logic control based on hybrid taguchi genetic algorithm for vehicle magneto-rheological suspensions

针对磁流变悬架的非线性以及动力学模型的不确定性,提出一种基于混合田口遗传算法的磁流变半主动悬架整车模糊控制策略。首先建立了基于磁流变减振器的整车动力学模型,并将车辆的振动控制分解垂向振动、俯仰、侧倾三个基本任务设计模糊控制器,进而设计了隶属函数和模糊控制规则;接着引入混合田口遗传算法实现对模糊控制器的隶属函数和模糊控制规则同时优化;最后进行实车道路试验来验证控制策略的有效性。试验结果表明,基于混合田口遗传算法的模糊控制能够降低确定路面激励下车身加速度峰峰值,降低随机路面激励下的加速度均方根值,显著提高车辆的平顺性,其控制效果要优于优化前的模糊控制策略。

To solve the problems of nonlinearity and dynamic uncertainty of magnetorheological(MR) suspension system,a fuzzy logic control strategy based on hybrid Taguchi genetic algorithm(HTGA) was proposed.As the first step,a full car dynamic model with four MR dampers was developed.The task of vibration control of vehicle was decomposed into three independent fuzzy control modules: a heave motion control module,a pitch motion control module and a roll motion control module.The membership functions and control rules of the three control modules were designed.Subsequently,HTGA was adopted to simultaneously tune the membership functions and control rules of the fuzzy controller.At last,road test was performed to validate the proposed control strategy.The results of the road test indicate that the MR suspension system with fuzzy controller tuned by HTGA can reduce greatly the peak-peak acceleration of the vertical vibration of vehicle under bump road excitation,achieve smaller root mean square of acceleration,and improve the ride comfort compared to the passive suspension system.Moreover,its control performance is superior to the simple fuzzy logic control strategy.