摘要
空气弹簧的刚度是可变的,采用常规的控制策略控制其刚度适时最优往往难以取得满意结果。神经网络可以用来表示任意复杂的非线性函数,模糊控制方法适应于时变和滞后系统的控制,本文即探索将这两种方法结合应用于半主动空气悬架刚度控制的可行性。首先建立1/4车辆动力学模型,设计模糊神经网络控制器和辨识器,然后基于Matlab对模型进行计算机仿真,并与被动悬架仿真数据进行比较。结果表明,采用模糊神经网络控制的半主动悬架振动性能明显优于被动悬架,进而提高了汽车平顺性和操纵稳定性。
The rigidity of air spring is variable,it is often difficult to get satisfactory optimal results of rigidity control by groovy control methods.Neural network is often used to analyze any complex nonlinear function,and fuzzy control method is mainly used to cope with hysteretic time-varying system.A feasibility study for air suspension rigidity control was carried out by using the two methods together.The quarter dynamics model of vehicle was established and a fuzzy neural network controller and an identifier were designed,and the simulation was carried out based on Matlab.Through comparing the simulation results of semi-active suspension with passive suspension,it can be known that the vibration performance of semi-active suspension with fuzzy neural network control is much better than that of passive suspension,which can improve ride comfort and maneuverable stability of a vehicle greatly.
出处
《汽车零部件》
2008年第1期42-45,共4页
Automobile Parts
基金
江西省自然科学基金:汽车半主动悬挂神经网络模糊控制技术的研究(0412029)
关键词
半主动空气悬架
刚度控制
仿真
模糊神经网络
Semi-active air suspension
Rigidity control
Simulation
Fuzzy neural network
