Stacking sequence blending of multiple composite laminates using genetic algorithms

• Published in: Composite structures Vol. 56; no. 1; pp. 53 - 62
• Main Authors: Soremekun, Grant, Gürdal, Zafer, Kassapoglou, Christos, Toni, Darryl
• Format: Journal Article
• Language: English
• Published: Oxford Elsevier Ltd 01.04.2002; Elsevier Science
• Subjects: Application of continuum mechanics to structures; Blending; Composite laminate; Cross-disciplinary physics: materials science; rheology; Dispersion-, fiber-, and platelet-reinforced metal-based composites; Exact sciences and technology; Fundamental areas of phenomenology (including applications); Genetic algorithm; Materials science; Optimal design; Other materials; Physics; Solid mechanics; Specific materials; Structural and continuum mechanics; Structural mechanics (beam, string...); Theory and numerical methods; Blending; Genetic algorithm; Composite laminate; Optimal design; Automation; Stacking sequence; Methodology; Computer software; Process planning; Layered materials; Continuous function; Experimental study; Sandwich structures; Genetic algorithms; Composite materials; Manufacturing cost
• ISSN: 0263-8223
• DOI: 10.1016/S0263-8223(01)00185-4

This paper presents a methodology for designing any two-dimensional array of laminated composite panels with varying loads such that continuity of some or all of the plies is maintained across adjacent panels. This design process is commonly referred to as blending and is beneficial in helping to minimize the manufacturing effort and cost of the panels while meeting structural requirements at minimum weight. The methodology discussed here involves an automated 2-step optimization process using a commercial genetic algorithm software code called DARWIN. The methodology is tested using a 3×3 array of sandwich panels, and an 18-panel group arranged in a horseshoe pattern. Results for each design problem are generated using ADOPTECH's Java-based composite laminate design software called OLGA, which utilizes the DARWIN optimization engine. A comparison between the individually optimized panels designs and the blended designs for each design problem is also presented.