摘要
针对汽车半主动悬架模糊控制器的模糊控制规则无法有效调整的问题,建立了两自由度1/4车辆模型.利用白噪声模拟路面激励并作为系统的输入,将人工神经网络与模糊逻辑控制相融合,采用人工神经网络模拟模糊控制过程,实现了模糊规则的自适应调整.将直接控制力作为参考控制力对神经网络进行训练,输出控制力结合开关控制策略实现悬架的半主动控制.仿真分析表明,神经模糊融合网络控制器相对于模糊控制器和被动悬架,使悬架性能得到了显著的改善.
Aiming at the problem that the fuzzy control rules of fuzzy controller can not be effectively adjusted, a quarter vehicle model with two degrees of freedom was established. The road excitation was simulated with the white noise and taken as the system input. The artificial neural network and fuzzy logic control were combined. In addition, the fuzzy control process was simulated with the artificial neural network, which could realize the adaptive adjustment of fuzzy rules. The direct control forces were taken as the reference control forces to train the neural network. Moreover, the output control forces were combined with the switch control strategy, which could realize the semi-active control of suspension. The simulated results indicate that compared with the fuzzy controller and passive suspension, the neural-fuzzy fusion network controller can obviously improve the performance of suspension.
出处
《沈阳工业大学学报》
EI
CAS
北大核心
2014年第2期170-175,共6页
Journal of Shenyang University of Technology
基金
辽宁省自然科学基金资助项目(201202161)
关键词
半主动悬架
神经模糊融合
减振器
乘坐舒适性
电控悬架
自适应控制
模糊逻辑控制
开关控制
semi-active suspension
neural-fuzzy fusion
shock absorber
ride comfort
electronic controlled suspension
adaptive control
fuzzy logic control
switch control
