The Reliability-based Optimum Design For Laminated Composite Structures Using Genetic Algorithms

Nowadays, the structural optimum design method joining with the structure system reliability analysis has achieved much progress. Reliability-based optimum design is regarded to be more reasonable than traditional optimum design, because the structure is considered as a whole, and the randomness in the period of its service is studied, that is to say some parameters are considered as random variables. In the reliability-based optimum design of laminate structures, the reliability demand of the laminate structure is generally combined into the object function of the optimum problem. The parameters of laminate structure are adjusted to get the maximal reliability with the restriction of the weight or fare. Such a structural optimum design based on the reliability can increase the design quantity and obtain obvious economic benefit, that is to say, the structure is both safe and economical. There are two kinds of failure criteria in composite laminates: the first ply failure (FPF) and the last ply failure (LPF). In practice, a laminate has considerable residual strength after the first ply failure. Therefore, the last ply failure stress is an important index when the laminate is used in the structures with severe light demand such as airplane. Moreover, it is important to evaluate the system (structure) reliability effectively based on the LPF criterion for those structures with defects and uncertainties in various parameters. The work of this paper is outlined as follows: 1. An overview is given of structure reliability analysis and reliability-based optimum design. The application of the reliability-based optimum design to composite laminates is addressed with several comments. 2. Some reliability analysis models of composite laminates are presented. The first order reliability method (FORM) is used to estimate the reliability of each ply group, and system reliability is computed based on both the first–ply failure (FPF) and the last-ply failure (LPF) assumptions. For the last-ply failure assumption, system reliability is estimated through the significant failure sequences. 3. The structural optimum design of composite laminates under probabilistic considerations is analyzed. On the foundation of reliability analysis, an optimum design problem with the system reliability index as the objective function is considered. The genetic algorithm (GA) is applied to the analysis of probabilistic optimum design of composite laminates. Numerical examples are worked out. By comparing the results based on the GA with those from the traditional optimization procedure, we can conclude that the GA method illustrates the simple flow, high compute efficiency and fast convergence in the optimum design of laminated composites.