双层自适应快速super twisting控制算法  

Adaptive dual layer fast super twisting control algorithm

作　　者：杨雅君[1] 廖瑛[1] 尹大伟[2] 郑宇昕[1] YANG Ya-jun;LIAO Ying;YIN Da-wei;ZHENG Yu-xin(College of Aerospace Science and Engineering, National University of Defense Technology, Changsha Hunan 410073, China;Naval Academy of Armament, Shanghai 200436, China)

机构地区：[1]国防科技大学航天科学与工程学院,湖南长沙410073 [2]海军装备研究院,上海200436

出　　处：《控制理论与应用》2016年第8期1119-1127,共9页Control Theory & Applications

基　　金：航天科技创新基金项目(CASC201502)资助~~

摘　　要：为提高super twising算法的收敛速度,解决现有算法存在的增益过估计问题,提出了两种自适应增益快速super twisting算法.分别通过快速终端滑模趋近律和增加线性项加快收敛速度.利用基于等效控制的双层自适应律调节增益,保证滑模存在条件的成立,同时使增益尽量的小.采用Lyapunov方法证明了新算法具有更优良的收敛特性,根据有界实引理和Schur补定理分别给出了两种算法的参数整定策略.仿真算例表明,在相同控制参数下,新算法的能耗与原算法接近,并具有更快的收敛速度和更强的鲁棒性.Two modified super twisting algorithms with adaptive gains are proposed for improving the convergence rate and averting the overestimation of gains which is often occur in present algorithm. One proposed algorithm is obtained by adding an additional linear term, and another algorithm is modified by using the fast terminal slide mode trending law.All variable gains in proposed algorithms are adaptively adjusted by using the dual layer adaptation schema which has exploited the concept of equivalent control; therefore, the gains are adjusted as small as possible. The proposed algorithms and dual layer adaptation are formally analyzed by using the Lyapunov methods to prove that the convergence properties are improved. The parameters setting strategies are proposed base on the bounded-real lemma and Schur complement lemma for both algorithms, respectively. Simulation results show that the convergence rate and robustness of proposed algorithms are superior to the original one when the total energy consumption is practically unchanged.

关 键 词：super twisting算法 收敛性 等效控制 自适应控制 滑模控制 观测器 

分 类 号：TP273.2[自动化与计算机技术—检测技术与自动化装置]