The residual stress generated during the curing process of composite structures will seriously reduce the material performance. This paper presents a two-stage genetic algorithm (GA) procedure to inversely determine the optimal molding parameters that minimize residual stresses. In our proposed two-stage GA procedure, a finite element model for Multiphysics simulation is first created to compute the residual stresses of the composite laminated plate for a given temperature curve. The FEM model is then modulated by an improved GA with the residual stresses of the plate as the objective function. The improved GA is called in two-stages: the first stage determines a set of likelihoods of the modeling parameters around which the "optimal" parameters may reside. The 2^nd stage zooms-in the areas centered by these likelihoods, which finds molding parameters that minimize the residual stresses. The results show that the proposed two-stage genetic algorithm is more efficient than the traditional genetic algorithm.

复合结构固化过程中产生的残余应力会严重降低材料性能。本文提出了一种两阶段遗传算法 (GA) 程序，以逆向确定最小化残余应力的最佳成型参数。在我们提出的两阶段 GA 程序中，首先创建用于多物理场仿真的有限元模型，以计算给定温度曲线下复合层压板的残余应力。FEM 模型然后由改进的 GA 调制，板的残余应力作为目标函数。改进的 GA 分两个阶段调用：第一阶段确定一组建模参数的似然性，“最佳”参数可能存在于这些似然性周围。在2^次阶段放大以这些可能性为中心的区域，从而找到最小化残余应力的成型参数。结果表明，所提出的两阶段遗传算法比传统遗传算法的效率更高。