基于遗传算法的疲劳裂纹扩展方法优化研究

Research on Optimization of Fatigue Crack Growth Method based on Genetic Algorithm

当前,用于预测疲劳裂纹扩展的方法多种多样,但无论哪一种裂纹扩展方法,都是以扩展点的扩展代替整个裂纹的扩展。因此,对扩展点的优化具有重要意义。考虑到遗传算法对多参数优化具有较好的效果,基于遗传算法对扩展点的个数和分布进行了优化研究;引进了“位置比”这个概念,以最外侧扩展点的位置比表征扩展点的分布;介绍了一种裂纹扩展的数值方法,计算数值结果与实验结果的误差,取该误差的倒数作为个体适应度。结果表明,当扩展点个数和最外侧扩展点位置比分别为11和0.95时,个体适应度最高,数值预测精度最好。

At present,there are various methods for predicting the fatigue crack growth,but no matter which kind of crack growth method,expansion of the expansion point replaces the entire crack expansion.Therefore,optimization of the expansion point is of great significance.Considering that genetic algorithm has good effect on multi-parameter optimization,this paper proposes to optimize the number and distribution of expansion points based on genetic algorithm.The concept of position ratio is introduced to represent the distribution of expansion points by the position ratio of the outermost expansion points.A numerical method for crack growth is introduced.The error between numerical results and experimental results is calculated,and the reciprocal of the error is taken as individual fitness.The results show that when the number of expansion points and the position ratio of the outermost expansion points are 11 and 0.95 respectively,the individual fitness is the highest and the numerical prediction accuracy is the best.