<title>Active vibration control of smart composite plates using self-adaptive neurocontroller</title>

Abstract

A neural network-based control system is developed for self- adapting vibration control of laminated plates with piezoelectric sensors and actuators. The conventional vibration control approaches are limited by the requirement of an explicit and often accurate identification of the system dynamics and subsequent 'offline' design of an optimal controller. The present study utilizes the powerful learning capabilities of neural networks to capture the structural dynamics and to evolve optimal control dynamics. A hybrid control system developed in this paper is comprised of a feed- forward neural network identifier and a dynamic diagonal recurrent neural network (DRNN) controller. Sensing and actuation are achieved using piezoelectric sensors and actuators. The performance of hybrid control system is tested by numerical simulation of composite plate with embedded piezoelectric actuators and sensors. Finite element equations of motion are developed based on shear deformation theory and implemented for a plate element. The dynamic effects of the mass and stiffness of the piezoelectric patches are considered in the model. Numerical results are presented for a flat plate. A robustness study including the effects of structural parameter variation and partial loss of sensor and actuator is performed. The hybrid control system is shown to perform effectively in all these cases.