【作者】 董小闵；

【导师】 黄尚廉； 李祖枢；

【作者基本信息】 重庆大学 ， 仪器科学与技术， 2006， 博士

【摘要】 随着高速公路网的大量建设,车速不断提高,人们对汽车的舒适性和安全性提出了更高的要求。悬架作为影响汽车性能的关键部件,采用能够根据路面情况和车辆运行工况进行实时控制的智能悬架是提高汽车性能的一条重要途径。基于磁流变技术的半主动悬架由于响应快,动态范围宽,功耗低,结构简单,成为了目前智能悬架的研究热点。磁流变阻尼器的研制和半主动控制策略的设计是磁流变半主动悬架的两项关键技术,随着磁流变阻尼器研制水平的不断推进,控制策略的设计与实现将成为磁流变悬架性能最终能否充分体现的关键因素。由于磁流变悬架是一个存在有不确定性、时滞的复杂非线性动力学系统,控制策略的设计极富挑战性,先前的一些半主动控制算法针对具体问题的特征,虽然都取得了一定的控制效果,但仍然存在局限性,较少从整车的角度对悬架进行协调解耦控制。基于此,本文以“仿人”,“仿生”为指导思想,以安装四支磁流变阻尼器的某国产轿车为研究对象,基于动觉智能图式仿人智能控制理论,运用理论分析、数值仿真和道路试验的方法,对磁流变悬架的建模,控制系统的设计与实现进行了研究,主要进行了以下研究工作:(1)建立了悬架的整车非线性动力学模型,对悬架动力学特性和模型误差进行了研究。整车动力学模型中磁流变阻尼器具有非线性特性;通过试验的方法测试了一些模型参数;研究了阻尼不断变化对悬架传递特性的影响;利用道路测试结果研究了整车悬架动力学模型的误差,分析了误差产生的原因,为控制算法的设计奠定了基础。(2)对磁流变减振器的工作特性进行了测试,建立了一种精确的磁流变阻尼器控制模型。根据汽车悬架减振器的标准和轿车的技术条件,利用电液伺服系统对阻尼器的示功特性、速度特性以及动态响应进行了测试,分析了磁流变阻尼器的工作特性;针对目前磁流变阻尼器控制模型存在的问题,建立了一种基于测试数据的简单、高精度的磁流变阻尼器控制模型。(3)基于整车动力学模型,提出了一种分姿态协调控制的仿人智能控制方法,设计了具有分层递阶的、多模态多控制器的仿人智能控制器。通过对复杂任务的分解,提出将汽车的运动姿态划分为八种进行分姿态协调解耦控制,分别设计了运行控制级和参数校正级。在运行控制级,设计了由八个动觉智能图式组成的图式群,直接对磁流变悬架进行分姿态协调解耦控制;在参数校正级,通过在线调节控制参数,算法具有较强的自适应性;最后运用混合田口遗传算法实现了对仿人智能控器多个控制参数的整定。(4)搭建了基于Matlab/Simulink的磁流变半主动悬架仿真控制研究平台。建更多还原

【Abstract】 With the rapid development of speedway and increase of vehicle velocity, people have made more demands on the comfort and security of automobiles. Suspension of vehicle is the key component that has great influence on performance of vehicle. The intelligent suspension that can improve ride comfort and stability of vehicle according to actual situation of real time is promising. In recent years, because of its rapid response, wide bandwidth, low consumption of energy and simple structure, the semi-active suspension based on magneto-rheological (MR) fluids has attracted much more attention in the research domain of intelligent structure. There are two key technologies including development of MR dampers and selection of control strategy. With the great progress made in some key technologies of MR dampers, the selection and implementation of control strategy have become crucial factors to make use of excellent MR suspension. Due to the nonlinearity, time-delay and uncertainty of MR suspension system, the design of control strategy is very challengeable. Though some semi-active control algorithms can achieve some control performances for particular characteristics, there are still some limitations that the coupling and harmony of full car suspension model are scarcely considered. Therefore, with the idea of human simulation and life-form simulation in this work, a saloon car made in China with four MR dampers is selected as research object, the modeling of MR suspension, the design and implementation of semi-active suspension system based on human simulated intelligent control (HSIC) theory are studied via theory analysis, numerical simulations and road tests. The main contributions of the dissertation include the following:(1) A full nonlinear dynamic car model considering heave vibration, pitch motion and roll motion has been established. The dynamic characteristic of suspension and the error of model are investigated. During modeling, nonlinearity of MR dampers is considered. Some model parameters of dynamical model are measured by experimental method. The effect of damping coefficient on transfer characteristic of MR suspension is discussed. The error of dynamical model is studied according to measuring results of actual road tests and the cause of error is analyzed also, with the purpose of providing the basis for the design of control algorithm.(2) An accurate control model of MR damper is established after the measurement of MR dampers’working characteristic. In accordance with national standard and the更多还原