钛合金动态本构模型参数敏感度分析及识别方法

发布时间：2018-05-09 16:25

  本文选题：钛合金 + 动态本构模型　； 参考：《四川大学学报(工程科学版)》2015年S2期

【摘要】：为分析钛合金动态本构模型参数的敏感度,克服传统单因素分析法的弊端,采用拉丁超立方抽样方法在整个参数空间中抽样,利用非参数统计方法中的Spearman秩相关分析法对本构参数随机输入样本集与其对应的目标函数输出结果集作相关性分析。建立了用Spearman秩相关系数等效求解参数敏感度的表达式,进而实现了参数敏感度的整体性分析,提高了参数敏感度分析结果的可靠性。基于参数敏感度分析结果,为减少本构模型参数识别的计算量,快速、精确、可靠地搜索最优解,建立了基于改进小生境算法、可疑峰值点判断策略和局域精确搜索技术的改进遗传算法。该方法克服了基本遗传算法存在的无法同时满足精度、可靠性和节省计算时间的缺陷,避免了容易产生早熟现象,增强了局部寻优能力。数值计算结果表明,提出的方法显著提高了遗传算法的全局搜索能力且计算精度高。提出的参数敏感度整体性分析方法和改进遗传算法不仅能很好地实现钛合金的动态本构参数识别,而且对其他工程材料本构参数的高精度识别具有重要参考价值。
[Abstract]:In order to analyze the sensitivity of the parameters of the dynamic constitutive model of titanium alloy and overcome the disadvantages of traditional single factor analysis, Latin hypercube sampling method was used to sample the whole parameter space. The Spearman rank correlation analysis method is used to analyze the correlation between the random input sample set of constitutive parameters and the output result set of its corresponding objective function. An expression for calculating the sensitivity of parameters by using Spearman rank correlation coefficient is established, which realizes the holistic analysis of the sensitivity of parameters, and improves the reliability of the results of the sensitivity analysis. Based on the results of parameter sensitivity analysis, in order to reduce the computational cost of parameter identification of constitutive model and search the optimal solution quickly, accurately and reliably, an improved niche algorithm is established. An improved genetic algorithm for judging suspected peak points and local precise search techniques. This method overcomes the shortcomings of the basic genetic algorithm which can not satisfy the accuracy, reliability and save calculation time at the same time, avoids the phenomenon of precocity and enhances the ability of local optimization. The numerical results show that the proposed method improves the global search ability of the genetic algorithm and has a high accuracy. The proposed method of parameter sensitivity integrity analysis and improved genetic algorithm can not only realize the identification of dynamic constitutive parameters of titanium alloy, but also have important reference value for high precision identification of constitutive parameters of other engineering materials.
【作者单位】： 四川大学建筑与环境学院;北京应用数学与力学研究所计算物理实验室;
【基金】：国家自然科学基金资助项目(B1520132013-1)
【分类号】：TG146.23
，

本文编号：1866715