基于主动悬架的模糊神经网络控制的研究

Active Suspension Based on Fuzzy Neural Network Control

传统的被动悬架由于阻尼参数不可任意选择和调节,减振性能不好,不能满足乘客的乘坐舒适性;而主动悬架可通过改变减振器的阻尼特性而适应不同的道路和行驶状况,改善乘坐舒适性和操纵稳定性;以汽车主动悬架为研究对象,建立了汽车二自由度1/4车体模型,提出了一种汽车主动悬架模糊神经网络控制方法,设计了模糊神经网络控制器;以B级路面作为随机路面输入,并利用Matlab进行仿真;通过动态仿真对被动悬架和主动悬架的特性进行了对比,仿真结果表明,该模糊神经网络控制器对车身加速度、悬架动挠度和轮胎动载荷都有很好的抑制。

The traditional passive suspension damping can not be arbitrarily selected and adjusted, the damping performance is not good. It can not meet passenger comfort. Active suspension by changing the damping characteristics of the shock absorbers adapt to different road and driving conditions, it will improve the ride comfort and steering stability. We study active suspension and establish a model of two--de- gree-freedom quarter vehicle. A new way of fuzzy neural network control of automobile active suspension is presented, and a fuzzy neural network controller is designed. Grade B road as a random road input, and use Matlab simulation. The result shows that the fuzzy neural net- work controller reduces body acceleration, suspension stroke and tire deflection.