半主动空气悬架的神经模糊控制  

Neural-Fuzzy Control Applied in Adjustable Volume Air Suspension with Additional Air Chambers

作　　者：江洪[1] 王子豪[1] 孔亮[1] JIANG Hong WANG Zi-hao KONG Liang(School of Mechanical Engineering, Jiangsu University, Zhenjiang 212013, China)

机构地区：[1]江苏大学机械工程学院,江苏镇江212013

出　　处：《重庆理工大学学报（自然科学）》2017年第3期1-11,共11页Journal of Chongqing University of Technology：Natural Science

基　　金：国家自然科学基金资助项目(51575241)

摘　　要：以带附加气室容积可调空气悬架的整车为研究对象,首先建立附加气室容积可调空气弹簧模型,并将该模型以弹簧力形式引入整车,然后设计以平顺性指标为主导地位的综合目标函数及约束条件。采用改进遗传算法对该目标函数进行逐段优化,同时得到不同工况下前悬架和后悬架模糊控制器的最优输出数据。最后将该输出数据作为导师信号供神经网络学习,从而建立整车半主动空气悬架的T-S型神经模糊控制器。仿真结果表明:在不同行驶工况下,相比被动空气悬架,采用神经模糊控制的半主动空气悬架的行驶平顺性显著提高,且满载时的改善效果优于空载。We studied on a whole vehicle equipped with adjustable volumes air suspension,firstly,a model of adjustable volumes air spring with additional air chamber was established,and then the model was substituted into a whole vehicle model in the form of the spring force. Secondly,we designed comprehensive objective function which regard the ride comfort indicator as leading status,and then the function of the whole vehicle was optimized cycle by cycle by the GA（ genetic algorithm） to receive the optimized output data of the fuzzy controller of front suspension and rear suspension under different driving conditions and then processed the data for learning of neural network to establish the T-S neural fuzzy inference systems. The results show that the semi-active airsuspension with fuzzy controller compared with the passive air suspension improved the ride comfort significantly under different driving conditions,and the improved effect was superior to vehicle which in empty load.

关 键 词：附加气室 空气悬架 神经模糊控制 改进遗传算法 平顺性 

分 类 号：U463.33[机械工程—车辆工程]