Optimum Design of a Smart Composite Beam by making Use of the Latest Control Theory

研究了最新控制理论在智能复合材料梁振动控制中的应用。主要研究结果如下：（1）制备了一种含有电流变液（ERF）和粘结压电陶瓷的碳纤维增强塑料（CFRP）梁，并对其在正弦激励下的振动特性进行了研究。利用模糊推理与神经网络的学习功能相结合的混合神经模糊系统，建立了含有强非线性执行器的被控对象的模糊模型。(2)在这种情况下，它是必要的设计模糊控制器的振动抑制的组合梁通过求解不等式满足稳定性条件，使用李雅普诺夫稳定性理论。但这些不等式不是严格意义上的线性不等式，求解起来非常繁琐。(3)通过将这些不等式转化为线性矩阵不等式，使得控制器的设计变得非常容易。此外，BIBO（有界输入有界输出）的稳定性可以证明。

The application of the latest control theory to vibration control of a smart composite beam was investigated. The results obtained are summarized as follows ;(1) A carbon fiber reinforced plastics (CFRP) beam including interleaved electro-rheological fluids (ERF) and bonded piezoceramics was prepared and oscillated under the forced sinusoidal external excitations. Fuzzy model of controlled element containing actuators with intensive nonlinearity was formed by using a hybrid neuro-fuzzy system combining fuzzy deduction with learning function of neural network.(2) In this case, it is necessary to design fuzzy controller for vibration suppression of the composite beam by solving the inequalities satisfying the stability conditions derived by the use of Lyapunov's stability theory. But these inequalities aren't linear in the strict sense and it is very tedious to solve them.(3) By transforming these inequalities to LMI (linear matrix inequalities), it became very easy to design controller. Moreover BIBO (bounded input bounded output) stability could be confirmed.