CONSTRUCTION OF SELF-TUNING SUSPENS10N SYSTEMS OF VEHICLES USING FUZZY THEORY

研究了利用模糊理论构建自整定车辆主动和半主动悬架系统的方法。该结构分为用于大型客车的主动悬架系统和用于小型乘用车的半主动悬架系统。在个人计算机上进行了仿真研究，并通过现场测试和实验证明了该方法的有效性。以车身的垂直和旋转加速度和速度为输入变量，通过选取模糊控制规则和隶属度函数，推导出主动悬架系统。利用乘积和重力法得到输出变量的去模糊化值。构造了一种能够根据道路输入的变化自适应调整控制结构的自调谐悬架系统，该系统的输出变量的隶属度函数随输入变量的变化而变化。提出了两种改善悬架性能的方法。首先是通过测量前轮提前输入的道路信息来构建预览控制。二是前后轴动态吸波器的构造。动态减震器利用前轮和后轮之间的石灰差产生的频率来减少振动。采用模糊控制规则，将悬架行程和时间增量作为输入变量，将阻尼力作为输出变量，推导出半主动悬架系统。由于阻尼力的选择至少需要10ms左右，并且会导致性能下降，因此考虑了时滞，构造了模糊控制规则。通过现场试验验证了半主动悬架系统的有效性，并通过四分之一汽车模型对气动主动悬架系统进行了实验验证。少

This investigation presents the construction of self-tuning active and semi-active suspension systems of vehicles using fuzzy theory. The construction is classified into active syspension systems for large-sized buses, and semi-active suspension systems for compact passenger cars. It is carried out by the simulation study using personal computers, and its effectiveness is shown by the field testing and the experiment. The active suspension system is derived by choosing the fuzzy control rules and the membership functions for the input and output variables where the vertical and rotary accelerations and velocities of the vehicle body are denoted as the input variables. The defuzzified value of the output variable is obtained by using the product-sum-gravity method. A self-tuning suspension system, which can adaptively tune the control structure according to the variation of the road input, is constructed where the membership function for the output variable is changed by the measurement … More of the input variable. Two kinds of approaches to improve the suspension performance is proposed. The first is the construction of preview control obtained by measuring the road input advanced from the front wheel. The second is the construction of dynamic absorbers fitting to the front and rear axes. The dynamic absorbers reduce the vibration with the frequencies generated by the Lime difference between the front and rear wheels. While, the semi-active suspension system is derived by choosing the fuzzy control rules where the suspension stroke and the time increment are denoted as the input variables, and the damping force is done as the output one. AS the selection of the damping force needs about 10 ms. at least and it leads the performance degradation, the fuzzy control rules are constructed by considering the time delay. The effectiveness of the semi-active suspension is shown by the field testing, and the pneumatic active suspension system is experimentally obtained by a quarter car model. Less