A sequential linear least square algorithm for tracking dynamic shapes of smart structures

Abstract

This paper presents a sequential linear least square algorithm for tracking dynamic shapes of piezoelectric smart structures. The dynamic shape discussed in this paper is defined as a host structural shape varying with time, and the tracking technique is to find an electric voltage history for each piezoelectric device over a time period so that the desired structural movements can be achieved. In the theoretical formulation, dynamic equations of piezoelectric smart structures are introduced by finite element analysis, and then a solution procedure for a set of time-dependent electric voltages is derived by combining the linear least square method and the Houbolt numerical integration scheme. The formulation indicates that this algorithm can be used to find the time-dependent voltages for tracking structural movements of piezoelectric smart structures. The present novel formulation is then demonstrated through numerical examples for tracking dynamic shapes of piezoelectric smart beams and plates. The numerical results for the smart beam are compared with the experimental ones. It is shown that the present sequential linear least square algorithm is capable of efficiently simulating dynamic shape tracking for smart structures. Copyright © 2006 John Wiley & Sons, Ltd.