Artificial neural networks for thermopiezoelectric systems

A radial basis function-artificial neural network modelling of thermopiezoe lectric systems is presented. The neural network model can emulate the elec trical response of two thermopiezoelectric layers bonded on a cantilever be am structure. The electrical outputs of thermopiezoelectric layers are due to sudden changes in temperature of thermo piezoelectric/beam system and ve rtical step force at the free end of the beam. The neural network is traine d so that it mimics the electrical response of the system for different the rmopiezoelectric layer locations. The test results of neural network are sh own together with the actual system results to illustrate the accuracy of t he network in predicting the thermopiezoelectric system behaviour to temper ature and force effects. Copyright (C) 1999 John Wiley & Sons, Ltd.