【作者】 张振华；

【导师】 杨明忠；

【作者基本信息】 武汉理工大学 ， 机械制造及其自动化， 2007， 硕士

【摘要】 随着汽车工业的高速发展，对汽车主动安全性和使用经济性、环保等方面提出越来越高的要求。汽车转向系统是控制其行驶路线和方向的重要装置，它直接影响汽车的操纵性和稳定性。对于汽车转向系统，除了要求其工作安全可靠、操纵轻便、机动性好、高效节能外，还要求它能够在各种工况(包括直线行驶、正常转向、快速转向、原地转向等)下，根据不同的行驶速度和路面状况，提供最佳的路感。目前，电动助力转向(简称EPS，Electric Power Steering)由于其集环保、节能、安全、舒适等性能为一体，正越来越受人们的关注。本文根据EPS系统的特点，以MATLAB作为主要仿真计算工具，应用经典控制理论和H_∞控制理论等，重点研究了EPS的鲁棒性控制问题。主要研究内容包括以下几个方面：(1)通过对EPS系统中助力电动机、扭矩传感器和齿轮齿条传动轴、方向盘等关键环节的分析，建立了EPS系统的动力学模型，并进一步变换得到了适合控制的状态方程模型。在建模过程中提出了实际研究对象存在非线性因素、模型参数摄动等不确定性因素，名义模型存在建模误差的事实。(2)考虑到EPS系统中存在的控制对象模型的不确定性和传感器噪声、路面干扰等因素，将H_∞控制理论引入EPS系统控制策略的研究中，通过建立更具一般意义的广义被控对象，根据性能要求选择合适的权函数，建立了广义状态空间，并利用MATLAB来对比H_∞控制与无H_∞控制两种策略的优劣。(3)在考虑系统的非线性部分的条件下，针对EPS系统参数摄动情况下的鲁棒控制器设计问题，利用现代控制有关线性不确定系统的相关理论，给出了EPS系统参数摄动情况下的鲁棒控制器，并利用MATLAB进行该闭环系统仿真，验证了其有效性。(4)在实际EPS控制系统中，由于存在参数不确定性以及执行器或传感器发生故障的不可避免性，应同时考虑二者对控制系统带来的影响。本文考虑了该系统对参数不确定性具有鲁棒性，同时针对传感器发生故障时具有完整性的容错反馈控制器，利用MATLAB进行闭环系统仿真。更多还原

【Abstract】 Along with the high-speed development of automobile industry, more and more requirements about automobile active safety, usage economy, environmental protection and etc. have been put forward. As important control equipments of its route and direction, automobile steering system affects its manipulate ability and stability directly. Besides work safety credibility, easy manipulation, good maneuverability, efficiently saving energy, more new requirements are put forward for automobile steering system, and it must provide the best road feeling under various conditions (such as straight drive, normal steering, fast steering and etc.) according to various drive velocities and road performances. Now the electric power-steering system (EPS) is paid more attention to by automobile industry, because it integrates the functions such as environmental protection, usage economy, drive safety and comfortable function...etc. In this thesis, according to the characteristics of EPS, the author researches its robust controller by means of classic control theory and H_∞control theory, and simulates its close loop system accord to MATAB.The following results are obtained:(1) After analyzing the following EPS components: assist motor, torque sensor, rack pinion shaft, steering wheel and so on, the dynamic EPS model is established, and according to the further transformation, the proper state-space model is obtained also. During the modeling process, it is put forward that there are modeling errors existed for the real object such as nonlinear factor and parametric uncertainty.(2) With the viewpoint of the model uncertainty, sensor noise, road disturbance and so on in EPS system, the H_∞control theory is introduced to EPS’s control strategy, and the general state-space model is established with the selected suitable weight function. The advantages of H_∞controller can be shown accord to MATLAB simulation.(3) The robust controller design of EPS with parametric uncertainty is considered under the nonlinear condition. The state feedback controller is designed by means of modern control theory, and close loop EPS system is simulated according to MATLAB. Simulating results show that the close loop EPS system with parametric uncertainties can be kept stable and has good responsiveness by this controller.(4) As parametric uncertainty and the failures of executers or sensors are inevitable in EPS system, their effects for the control system should be considered. The failure -tolerant controller is robust to the parametric uncertainty and can tolerate the failures of sensors. According to MATLAB, close loop EPS system is also simulated.更多还原