Research On Key Technologies Of Active Vibration Control For Piezoelectric Smart Structures

Piezoelectric materials can be sensor and actuator in smart structures because of their good piezoelectric and electromechanical coupling features. Active vibration control system based on piezoelectric materials has strong environmental adaptability, flexibility and wide controllable frequency, fast response speed, so it has been widely studied in the field of aerospace, mechanical, traffic, etc. At present, many achievements about the active vibration control of piezoelectric smart structure produced, but there are many key problems need to be solved. Five aspects are mainly studied from piezoelectric device optimization integrated the control law, design of robust control algorithm considering external mass perturbation, design of improved adaptive control algorithm considering saturation restrain under the nonlinear disturbance, local strain compensation in collocated vibration control and fault tolerant vibration control. The main works and innovative results in this dissertation are as follows:(1) Electromechanical coupling model of piezoelectric smart cantilever beam is established based on the Hamilton principle and modal transformation theory. Particle Swarm Optimization(PSO) algorithm was used to carry out the simultaneous optimization for the arrangement of sensors and actuators as well as the feedback control gains. A multiple performance index with the minimum storage energy and control energy as well as the fastest vibration decay was selected as an optimum objective function. The optimal piezoelectric patch positions and control gains at specific vibration modes were found out. In addition, the influence of both length and number of piezoelectric patches on the control effect was analyzed. The simulated results showed that the control effect for structure vibration was obvious, and the feasibility of simultaneous optimization based on particle swarm method was proved.(2)Aiming at modeling uncertainty and disturbance of piezoelectric smart structure,a new robust model reference control law based on Lyapunov method was designed. The controller included a state feedback controller for ensuring stability and a robust compensator for compensating uncertain effects. The control effect was analyzed based on three different compensation functions under single mode or multi-modes, and the possible chattering caused by sign function could be avoided by using tanh or erf function. Simulation and experimental results showed that the proposed method has good control performance and robustness under external mass disturbance.(3)An improved minimal control synthesis(LQR-IMCS) algorithm was put forward by Popov standards for suppressing the multi-modal nonlinear vibration of stiffened panel on the basis of LQR and traditional MCS method. It was consisted of adaptive feedforward factor, adaptive feedback factor and compensator factor. Simulation and experimental results showed that this method can effectively avoid saturation of adaptive controller and have better effect and robustness than traditional MCS under sine, impulse and random signal disturbances.(4)In order to overcome the problems resulting from the local strain in the collocated configuration of piezoelectric structure and the compensation damping ratio of the positive position feedback(PPF) algorithm, a new method was proposed which used a low-pass filter to attenuate the local high-frequency strain mode and employed the direct velocity negative feedback(DVFB) to modify the PPF algorithm for the damping ratio improvement. The mechanism and influence factors of local strain were analyzed through the harmonic response for a thin plate by using ANSYS. The conditions for the stability of the improved algorithm were analyzed according to the Royce criterion. Simulation and experimental results showed that,under either sine or random signal excitation, the method integrating low-pass filtering, PPF and DVFB can greatly reduce the mode amplitude at 90 and 162 Hz and effectively suppress the harmonic component during the single PPF experiment, with a simultaneous control effect of the two modes being up to more than 10 d B.(5) Aiming at the reliability problems of vibration control system in flexible beam, wavelet packet and RBF neural network was firstly put forward to identify the faults of piezoelectric devices, then two fault tolerant control methods were designed based on PPF-DVFB algorithms for typical piezoelectric patches fault, and good effect of fault tolerant control was lastly proved by experiment.