基于有限元和系统辨识的智能结构主动控制  

Active vibration control of smart structures based on FEM and system identification technique

作　　者：董兴建[1] 刘龙[1] 孟光[1] 叶林[2] 

机构地区：[1]上海交通大学振动冲击噪声国家重点实验室,上海200030 [2]悉尼大学宇航机械与机电工程学院

出　　处：《振动工程学报》2005年第3期329-334,共6页Journal of Vibration Engineering

基　　金：国家自然科学基金项目(50390063);国防科技重点实验室基金资助项目(51463040403JW0301)

摘　　要：探讨了一种压电智能结构的设计方法,包括动力学建模、控制器设计和闭环系统有限元仿真。首先采用有限元方法计算滤过白噪声激励下压电智能结构的响应,以此响应作为系统辨识方法的输入,采用基于观测器/K a lm an滤波器的系统辨识方法(O bserver/K a lm an filter iden tification,OK ID)得到系统的M arkov参数,亦即单位脉冲响应的采样值,然后采用特征系统实现算法(E igensystem R ea lization A lgorithm,ERA)得到系统的最小实现,基于此模型采用LQG优化算法设计鲁棒控制器,并将反馈控制引入有限元模型进行闭环系统仿真,根据仿真结果评价设计方案。此方法克服了有限元模型无法直接用于控制器设计的缺点,通过将反馈控制引入有限元模型,可用有限元方法研究控制器的性能,也适用于设计其它复杂智能结构。The objective of this study is to develop a general design scheme for actively controlled piezoelectric smart structures. The scheme involves dynamic modeling, designing control laws and closed-loop simulation in a finite element environment. Based on the structure responses determined by finite element method, a modern system identification technique known as Observe/Kalman filter Identification （OKID） technique is used to determine the system Markov parameters. The Eigensystem Realization Algorithm （ERA） is then employed to develop an explicit state-space model of the equivalent linear system for control law design. The Linear Quadratic Gaussian （LQG） control law design technique is employed to design a control law. The finite element model is then modified to accept control laws and perform closed-loop simulations. Finally, numerical examples have demonstrated the validity and efficiency of proposed design scheme. Without any further modifications, the design scheme can be readily applied to other complex smart structures.

关 键 词：压电材料 智能结构 特征系统实现算法 辨识 振动控制 

分 类 号：O32[理学—一般力学与力学基础] TB123[理学—力学]