An efficient approach for stacking sequence optimization of symmetrical laminated composite cylindrical shells based on a genetic algorithm. (September 2019)

• Record Type: Journal Article
• Authors: Wei, Ranfeng; Pan, Guang; Jiang, Jun; Shen, Kechun; Lyu, Da
• Is Part Of: Thin-walled structures. Volume 142(2019)
• Page Start: 160
• Page End: 170
• Publication Date: 2019-09
• Subjects: Composite cylindrical shell -- Buckling pressure -- Stiffness coefficient -- Genetic algorithm -- Finite element method -- Optimization; Thin-walled structures -- Periodicals; 690.1
• DOI: 10.1016/j.tws.2019.05.010
• Languages: English
• ISSNs: 0263-8231

Abstract

Abstract: This paper is devoted to solving the stacking sequence optimization problem of symmetrical laminated composite cylindrical shells subjected to hydrostatic pressure. First, a conventional genetic algorithm (GA) coupled with a finite element analysis optimization method is developed to search for the best laminations with the maximum buckling pressure. These optimal laminations share similar extensional stiffness coefficient ratios A 11 / A 22 and bending stiffness coefficient ratios D 11 / D 22 because the two ratios of the optimal lamination fluctuate slightly around a specific value. Based on this phenomenon, a stiffness coefficient-based method (SCBM) is then proposed. The method is integrated with the GA and the stiffness coefficient calculation to identify the lamination whose stiffness coefficient ratios ( A 11 / A 22 and D 11 / D 22 ) are closest to those obtained previously. The effectiveness of the SCBM is validated by comparison with the optimal results. The proposed method is then extended to more complex symmetrical laminations. Finite element analysis is also coupled with the GA as a control group. Comparisons reveal that the two methods lead to similar characteristic lamination patterns and maximum buckling pressures, which suggests that the SCBM works well for complex laminations. Moreover, the SCBM is found to be significantly more efficient because it only needs to calculate the stiffness coefficients rather than analyse the entire structure during the optimization. Highlights: The stiffness coefficient ratios of the optimal laminations are found to fluctuate slightly around a specific value. The SCBM is proposed to find a lamination with stiffness coefficient ratios closest to this specific value. The proposed SCBM proved to be more efficient than the most commonly used GA-FEM.