【作者】 裴福俊；

【导师】 方一鸣；

【作者基本信息】 燕山大学 ， 控制理论与控制工程， 2003， 硕士

【摘要】 本课题是在已完成的国家“九五”科技攻关项目“结晶器非正弦振动的研究”的基础上，针对该项目中的液压伺服驱动连铸结晶器振动模拟装置所做的控制方面的研究。由于液压伺服驱动位置系统中存在参数摄动、外干扰和非线性摩擦力等问题，而模糊控制不需要精确的数学模型，神经网络可以有效地解决非线性问题，因此本文采用模糊控制和神经网络等控制方法对液压伺服位置系统进行控制，具有较大的实际意义和理论价值。 本文首先利用已得到的液压伺服位置系统被控对象的简化数学模型作为控制器设计的依据。在对模糊控制和神经网络理论的特点和应用等进行概述的基础上，分别设计了模糊滑模控制器、基于模糊自学习的滑模控制器、基于改进Elman网络辨识的模糊神经网络控制器、模糊神经网络滑模控制器，以及针对液压伺服系统的非线性机理模型设计了模糊神经网络滑模控制器。其中，针对液压伺服系统非线性机理模型的模糊神经网络滑模控制方法，是以对伺服阀和液压缸的非线性函数的分析得出的状态方程为基础，并结合Lyapunov函数方法设计了控制器，从而有效解决了液压伺服系统的非线性问题。通过对所设计控制器的仿真研究表明，本文所设计的液压伺服位置控制系统具有较好的动静态性能和较高的跟踪精度，证明了所设计控制器的可行性。更多还原

【Abstract】 Based on the having been finished project of the state "the ninth five years" tackling key scientific and technological problems, this paper makes some detailed research on the continuous casting analogy device in the laboratory of our school. As far as the hydraulic servo position system is concerned, there are parameters variety, outside disturbance and the nonlinear friction in it. In fuzzy control method, it is well known that accurate model of the system is not needed. Moreover, neural network can solve the problems caused by non-linear effectively. As a result, in this paper, the two intelligent control methods are combined to control the hydraulic servo position system, and that has practical meaning and theory value.In this paper, the simplified models derived from the hydraulic system are used to design the controllers. First the summary of the characteristics and application of fuzzy control and neural network theory are narrated. Then, the fuzzy sliding mode controller, the sliding mode controller based on fuzzy self-learning, fuzzy neural network controller based on the modified Elman neural network identification, the fuzzy neural network sliding mode controller and a fuzzy neural network sliding mode controller based on the hydraulic mechanical model are proposed. Where, with Lyapunov function and the state equation that is analyzed from the nonlinear function of servo valve and hydraulic cylinder, the fuzzy-neural network sliding mode controller based on the mechanical model of hydraulic servo system is designed. Consequently, This controller is availability to the nonlinear problem existing in the hydraulic system. At last, The simulation shows that hydraulic servo position control system has good performances in the dynamic and static state control accuracy, thus the feasibility of these control methods are proved effectively.更多还原