Research On Intelligent Active Vibration Control Of Photostrictive Laminated Flexible Plates And Shells Structure

The plate and shell structures have a broad application backgroud in the area of aviation and aerospace. The shape control and vibration control of this kind of structures have always been major and difficult issues in the system design and engineering application. At present, it is an effective way by using smart materials to provide effective actuation for the purpose of achieving vibration suppression of plate and shell structures. The conventional smart materials activated by electric or magnetic signals require complex accessorial devices against lightweight and small size of system. Additionally, they also require connecting hard-wire to transimit energy sources and control commands. The hard-wire signal transmission busses can easily attract electromagnetic noise disturbance to affect the control signal transmission accuracy and real-time performance. The new smart material, lanthanum modified lead zirconate titanate(PLZT), can directly transform light energy to mechanical energy and is not affected by electromagnetic interference. PLZT actuators are suitable for the non-contact actuation and remote control in the space environment. In this dissertation, the intelligent active vibration control of plate and shell structures laminated with PLZT actuators is studied, which is mainly including four aspects of theoretical research, namely, the actuator configuratioin, dynamic modeling, actuator location optimization and intelligent active control algorithms for independent mode and multiple ones.The main work and novel research performed in this dissertation include:(1) Based on the constitutive equations of PLZT actuator in view of light-thermal-elastic-electric coupling, the main factors influencing the performance of PLZT actuator are analyzed by the numerical simulation method. The current commonly used actuator configurations are analysed and compared; From such an analysis, it is concluded that existing actuator configurations under the external light source can only elongate but not shorten, therefore can only produce a one-way membrane control force; Furthermore, two kinds of multi-piece combination actuator configurations to be able to produce both positive and negative membrane control force are been proposed to remedy this deficiency of the existing actuator configuration.(2) Based on the plate and shell structure vibration theory, the general dynamic model of photo-electric laminated plate and shell structure are established, which can be applied to plate and shell structures with different structure types and geometric parameters. Using this model,system dynamics equations for different types of photo-electric laminated plate and shell structures can be further deduced, such as rectangular plate, cylindrical shell, spherical shell and conical shell, etc. Based on dynamic model, the modal control equations of photo-electric laminated plate and shell structures are established using the modal expansion technique.(3) With considering the switching actuator and nonlinear driving characteristics of PLZT actuator, independent modal variable structure fuzzy controller is proposed. Comparing with Lyapunov control, namely constant light intensity, and velocity proportional feedback control, namely variable light intensity, the proposed control scheme has two aspects of advantage, that is, on the one hand, switch function of light illumination direction is optimized, and the optimal switch function of light illumination direction is obtained; on the other hand, light intensity quantitative factor self-adjusting fuzzy controller is proposed with giving full consideration to the driving characteristics of of PLZT actuator. The proposed control scheme combines the advantages of fuzzy control and variable structure control. It is not dependent on the system accurate models, and overcomes nonlinear driving characteristics of PLZT actuator. Its control effect is much better than velocity feedback control.(4) By the practical example, the change rule for the modal control factor of corresponding controlled modes is given when actuators’ position changes. It is included that there is one or more of extremum areas for the corresponding mode. In the extremum areas, the amplitude of actuator modal control factor is obviously greater than the other patch area. Moreover, with modal half wave numbers change, the distribution of the extremum region also changes. Furthermore, in order to suppress multiple controlled modal vibrations simultaneously, actuators are arranged on location to have as large as possible modal factor on all controlled modes. In order to resolve this problem, genetic optimal algorithm of actuator placement of plate and shell structures is proposed, in which the controlled modal control factor is selected as the optimization objective function, and actuator location coordinates are taken as optimization variables. Taking plate and shell structure laminated with the proposed actuator configuration for eaxmple, actuator location optimization of the multi-modal vibration is carried out, and optimization layout locations of plate and shell structures are given under the corresponding actuator configurations.(5) Aiming at multi-modal active vibration control of photo-electric laminated plate and shell structure, optimal fuzzy control algorithm is proposed. The optimal fuzzy control algorithm is combined with the current mature optimal control and fuzzy control, In this algorithm design process, system control and actuator control are separately considered, and its design step is divided into two steps: firstly, LQR control law based on the simplified system model is designed, and then by means of the fuzzy controller to adjust the photo-electric actuator input light intensity makes photo-induced strain to approximate the optimal control law; The method solves the contradiction that linear system control methods cannot be directly applied in photo-electric laminated system. It is by decomposing a complicated problem into a set of simplex problems that simplifies the design of the controller and multi-modal active vibration control of photo-electric laminated plate and shell structure is achieving successfully. Combined with the presented optimal fuzzy algorithm, effectiveness of the actuator placement optimization criterion function of multi-modal vibration is verified through simulation comparison.(6) Multi-modal active vibration algorithms of fuzzy neural network control(FNNC)and self-organizing fuzzy sliding mode control(SOFSMC)are proposed, in which actuator control and structure system are considered as a whole. The presented fuzzy neural network controller utilizes the advantages of fuzzy control and neural network control. In order to simplify system structure, the proposed fuzzy neural network controller is based on RBF neural network structure, and adopted two inputs and single output structure. However, in multi-modal vibration control, the number of controlled variables is more than input one of the controller, to resolve this problem, two level sliding mode surface used in underactuated control theory is taken into account, first of all, the controlled modal displacement and its velocity signal are combinated linearly to construct the first-level sliding mode function, and then linear combination of all first-level sliding mode function constructs the second-level sliding mode function; The second-level sliding mode function and its derivative are used as a FNNC input variables. FNNC active controller constructed with neural network is not dependent on the system mathematical model, and has the online learning ability of fuzzy rules and membership functions.(7) Self-organizing fuzzy sliding mode active vibration control algorithm is proposed. In which, it is by introducing a second-level sliding mode function in the controller design process that makes the system control order to be reduced and simplifies structure of fuzzy control system; it is by introducing self-organizing learning algorithm that makes the controller be with on-line self-learning ability and overcomes the disadvantage the conventional fuzzy sliding mode controller is dependent on the system rules; the use of fuzzy sliding mode in system control softens control signal, and reduces or avoids the chattering phenomenon of conventional sliding mode control; singleton fuzzy rule parameters are used, which is automatically adjusted with self-organizing learning algorithm; the used self-organization learning algorithm is different with one of the current published literature, the new multi-modal active vibration control self-organizing learning algorithm is based on linear auto-regression moving average model of multi-modal vibration system of photo-electric laminated structure In order to verify the effectiveness of the proposed intelligent active control algorithms, numerical simulation for multi-modal active vibration control of plate and shell structures are carried out.