基于遗传算法和梯度算法压电层合结构的最优形状控制  

OPTIMAL SHAPE CONTROL OF MULTILAYERED PIEZO-STRUCTURES BASED ON THE GENETIC ALGORITHM AND GRADIENT METHOD

作　　者：王建国[1] 丁根芳[1] 覃艳[1] 

机构地区：[1]合肥工业大学土木建筑工程学院,合肥230009

出　　处：《固体力学学报》2008年第1期59-65,共7页Chinese Journal of Solid Mechanics

基　　金：国家自然科学基金(10572046);高等学校博士点专项科研基金(20030359005);安徽省自然科学基金(050440504)资助

摘　　要：基于压电层合结构的有限元方程,运用ANSYS/APDL语言,编制了力-电耦合有限元分析程序(MPFEMP).以该程序为计算基础,采用遗传算法和一阶梯度优化算法,以压电片尺寸为设计变量,以压电层合梁和板的预期位移或最小重量为目标函数,给定初始变量和适应度函数,通过循环迭代MPFEMP计算程序,研究了多点控制的压电层合梁板结构的形状最优控制.结论对比分析证明了两种优化方法分析压电层合结构的有效性,同时,对复杂多层智能结构的最优形状控制和主动控制研究具有一定的参考价值.Based on the finite element formulation of piezoelectric laminated structures, a code on electromechanical coupling has been developed in the paper, where the genetic algorithm and first-order gradient optimal technique have been adopted in the optimization with piezo size and voltages as two design parameters. The expected displacements or minimum weights of piezoelectric lamina are taken as the objective function. By executing the mechanical piezoelastic finite element method program（MPFEMP） program circularly, the optimal shape control of a piezoelectric lamina has been studied in detail with multiple control reference point.. Numerical results show that the genetic algorithm and the gradient optimal method are very valid in the optimal designing and shape controlling of smart structures.

关 键 词：压电层合结构 有限元 形状最优控制 遗传算法 梯度优化算法 

分 类 号：O342[理学—固体力学]