【作者】 赵永胜；

【导师】 陈立平； 张云清；

【作者基本信息】 华中科技大学 ， 机械设计及理论， 2007， 博士

【摘要】 随着科技和生产的迅速发展,所研究的机电系统越来越复杂,对自动化控制的控制精度、响应速度、系统的稳定性与适应能力的要求也越来越高。然而,由于机电系统存在强非线性、外部扰动、参数时变等因素的影响,采用常规控制算法实现机电系统控制时,难以满足其不断提高的性能要求。针对机电控制系统的上述问题,本论文以滑模变结构控制理论为主要架构,引入了模糊自适应系统实现滑模变结构控制器的参数在线自适应调整,并以Lyapunov准则为控制系统稳定性分析的理论基础,分别以提高控制系统的实时性和鲁棒性为目标,设计了两种控制器来处理上述问题。针对模糊滑模控制器(FSMC)的实时性问题,设计了增强型模糊滑模控制器(EFSMC)。在确保被控非线性系统的全局渐进稳定基础上,提出了以局部线性化模糊自适应调节系统调整滑模控制器增益的控制方法,使得系统状态在快速趋近被控系统滑模面的同时,模糊自适应系统的模糊规则大大减少,提高了控制系统计算效率,从而保证了整个控制系统的实时性要求。另外,以进一步提高控制系统鲁棒性,降低抖振为目的,本论文还提出了模糊自适应动态滑模控制器(FADSMC),实现对被控系统的有效控制。实际系统中,由于存在系统不确定性、外部扰动以及量测噪声等不利因素,采用常规控制器难以获得平滑的控制输出,尤其是对一类要求输出控制力及其变化率作为伺服系统目标输入的被控系统,以(U(k)-U(k-1))/Δt=ΔU来计算控制力变化率的方式,将导致变化率输出产生较大的波动,从而造成被控系统的不稳定。而动态滑模控制器能够直接输出控制力的变化率,且控制力可通过对控制力变化率进行积分来求得,因而输出的控制力较为平滑,滑模控制器的抖振现象能够得到有效缓解。同时,模糊自适应调节系统的采用也使得该控制器具有快速反应能力,进一步提高了控制系统的鲁棒性。本论文利用三个模拟系统以及一个实验系统,来验证所提出的控制器性能。分别是:(1)以增强型模糊滑模控制器控制二自由度机械手和三自由度并行机械手模拟系统,来验证所提出增强型模糊滑模控制器的有效性与实时性。(2)以模糊自适应动态滑模控制器实现对1/4车主动悬架系统的有效控制,与被动悬架系统、线性二次型Gauss最优控制(LQG)以及常规滑模控制系统(CSMC)控制的主动悬架系统相比较,仿真实验结果表明该主动悬架系统的鲁棒性得到有效提高。(3)建立基于dSPACE的AMT自动离合器快速原型实验系统来验证所提出的增强型模糊滑模控制器和模糊自适应动态滑模控制器的有效性。所有模拟仿真和实验结果表明,所提出的控制器具有良好的控制性能。更多还原

【Abstract】 The electromechanical system is more and more complicated with the rapid development of science and technique, and the requirement of precision, respond time, stability and adaptive ability is also improved. However, due to the non-linear dynamic characteristic of electromechanical system, external disturbance and parameter uncertainty, it is difficult to be control precisely such nonlinear-system when adopting the conventional control arithmetic. In this paper, two control arithmetics based on the sliding mode controller are proposed for the purpose of improving the robustness and real-time requirement, respectively. The sliding and global stability of these two controllers are approved in terms of Lyapunov full quadratic form. The fuzzy logic adaptive system is introduced to tune the parameters of sliding mode controller. The performance of controlled system is improved effectively by adopting the proposed fuzzy sliding mode controllers.However, the expensive computing requirement of the complicated algorithm in fuzzy sliding mode controller (FSMC) may limit its on-line application to the complicated electromechanical system. Therefore, the enhanced fuzzy sliding mode controller (EFSMC) is proposed to improve the real-time requirement of the non-linear controlled system. The region-wise linear fuzzy adaptive system is adopted to tune the gain of the sliding mode controller, which can force the states of controlled system hit the sliding mode surface, quickly. The number of the fuzzy rules for the fuzzy logic adaptive tuner is also reduced. As a result, the complexity of FSMC is reduced, which ensure the real-time requirement of the controlled system.In addition, in order to improve the robustness of the controlled system and alleviate the chattering phenomenon, the dynamic sliding mode controller with fuzzy adaptive tuning (FADSMC) is proposed in this paper. For a real controlled system, the external disturbance and measuring noise must be considered during the design of the controller. Therefore, it is difficult to obtain the smooth control output for the conventional controller. Especially, for a class of controlled system which require the control force and its change as the control input of the servo system, the discontinuous jump will occur when the (U(k)-U(k-1))/Δt=ΔU is adopted to compute the change of control force. Then, the controlled system will not work stably and the robustness and control precision is reduced. However, the change of control force can be obtained by adopting the dynamic sliding mode controller. The control force can be easily obtained from the integral of its change. A smoother control force can be obtained and the chattering phenomenon is also alleviated. The fuzzy adaptive system is introduced to improve the robustness and stability of the controlled system.In this paper, three simulation systems and an experiment system are adopted to demonstrate the proposed controllers. (1) A 2 degree of freedom (DOF) polar manipulator and a 3-DOF parallel manipulator simulation system have been carried out to demonstrate the performance of the proposed the proposed EFSMC controller. (2) The proposed FADSMC controller is applied in the active suspension system to improve its performance. The simulation results are compared with that of the passive suspension, LQG control active suspension and CSMC control active suspension system. (3) The rapid prototype experiment of the automated clutch of AMT based on the dSPACE is designed to demonstrate the performance of the proposed EFSMC and FADSMC. The simulation and experimental results show that the better performance can be obtained by adopting the proposed controllers.更多还原