中文摘要：

以热轧立辊为研究背景，针对其液压伺服系统存在的非线性、参数不确定性以及负载干扰等特点，基于模糊基函数网络提出一种自适应控制方法。首先将非线性系统线性化并将其作为已知系统，利用这部分已知的动态特性设计反馈控制使标称系统稳定。然后利用模糊基函数网络仅学习非线性系统不确定性的上界，将输出作为补偿控制器的参数，并在Lyapunov稳定意义下构造自适应控制器，该自适应控制器不仅确保了闭环系统的鲁棒性而且加快了跟踪误差的收敛速度。将该控制器应用于某热轧立辊电液位置伺服系统中进行仿真研究，结果表明，该控制器优于传统的PID控制器，可以取得较好的控制效果。

英文摘要：

An application of an adaptive fuzzy system for compensating the effects induced by the nonlinear and parameter uncertainty existed in the electro-hydraulic servo system is presented. Firstly, the system is divided into two parts, one is known linear part and the other is unknown nonlinear part. A feedback controller is designed for the known part to assure it stable. Secondly a fuzzy basic function is used only to study the upper bound of the unknown part and the output is used as the parameter of compensating controller. Finally an adaptive fuzzy controller and its adaptive parameters are designed based on Lyapunov function to insure the closed system＇s robustness and the convergence to zero is speeded up. The proposed scheme is implemented and tested on the electro-hydraulic position servo system of vertical roller of hot strip mill. The results show the proposed controller can achieve better performance than conventional PID one.