A modified differential evolution algorithm for tensegrity structures

详细信息    查看全文

• 作者：Dieu T.T. Do ^{thanhdieu0801@gmail.com" class="auth_mail" title="E-mail the corresponding author} ; Seunghye Lee ^{lsh2002437@gmail.com" class="auth_mail" title="E-mail the corresponding author} ; Jaehong Lee ; ^{jhlee@sejong.ac.kr" class="auth_mail" title="E-mail the corresponding author}
• 关键词：Form-finding ; Tensegrity structure ; Force density method ; Differential evolution (DE) ; Modified differential evolution (mDE) ; Genetic algorithm (GA)
• 刊名：Composite Structures
• 出版年：2016
• 出版时间：15 December 2016
• 年：2016
• 卷：158
• 期：Complete
• 页码：11-19
• 全文大小：815 K

文摘

In this paper, a novel modified differential evolution (mDE) algorithm for advanced form-finding of tensegrity structures is proposed to define an appropriate candidate for strut members. The form-finding process only requires topology and member type of a tensegrity based on the force density method. The proposed algorithm improved from original differential evolution (DE) is performed to reduce significant computational cost. In the mDE, scale factor F and crossover rate c are adjusted as well as the mutation and selection phases of the original DE are also replaced by the best individual-based mutation and elitist selection techniques. The objective function of the product of α and β related to eigenvalues and force densities is minimized. Since force density values are considered as continuous design variables, optimal solutions obtained by mDE are more accurate than those solved from discrete design variables of GA. Several benchmark numerical examples of two- and three-dimensional tensegrity structures are investigated to verify the effectiveness and robustness of the proposed algorithm by comparing obtained results with those of other methods in the literature.