Multiobjective Genetic Algorithm Approach to the Optimization of a Rotating Flexible Arm With ACLD Patch

Abstract

This paper describes the use of multi-objective genetic algorithm (MOGA) to solve an integrated optimization problem of a rotating flexible arm with Active Constrained Layer Damping (ACLD) treatment. The arm is rotating in a horizontal plane with triangular velocity profiles. The ACLD patch is placed at the clamped end of the arm. The design objectives are to minimize the total treatment weight, the control voltage and the tip displacement of the arm, as well as to maximize the passive damping characteristic of the arm. Design variables include the control gains, the maximum angular velocity, the shear modulus of the viscoelastic layer, the thickness of the piezoelectric constraining and viscoelastic layers, and the length of the ACLD patch. The finite element method, in conjunction with the Golla-Hughes-McTavish (GHM) method, is employed to model the ACLD flexible arm to predict its dynamic behavior, in which the effects of centrifugal stiffening due to the rotation are taken into account. Reasonable Pareto solutions are successfully obtained. It is shown that MOGA is applicable to the present integrated optimization problem.