Neural network control of elastic waves transformations and suppression of amplitudes of oscillations in laminated piezoelectric composites

摘要

The converse piezoelectric effect is used to suppress the amplitudes or to modify the frequency of elastic waves which propagated along thickness of laminated metal-ceramic plate when on the face surface of a plate is applied the oscillating pressure. The control problems involve the minimisation of quadratic cost functional depending on the velocity of vibration of the back surface of the plate and functional depending on the displacement of the back surface by using the adjustment of the neural network with the input signals as the voltages of sensors piezoelectric layers and the output signals as voltages applied to the actuators PZT-ceramic layers. The presented results are concerned with computer simulation of the proposed animated composite materials as joint mechanic, neural network and electronic system (mechatronic system with scale level of the order 10{sup}(-3)÷10{sup}(-4) meter and deformational waves interactions), which has been named as the Matrix Electronic Materials (MEM; The results of modelling of elastic waves transformations have been shown an ability of MEM to suppress on the order the amplitudes of oscillations.