基于子结构的有限元模型修正方法

摘要

This paper proposes a substructure-based model updating method.The global structure was divided into independent manageable substructures, which were analyzed to calculate the substructural eigensolutions and eigensensitivity. Afterwards, the independent substructures were constrained to calculate the eigensolutions and eigensensitivity of the global structure.Finally, the eigensolutions were used for the objective function and the eigensensitivity was used to indicate the searching direction to achieve the model updating.When the local area of a structure was changed,only one or more substructures were analyzed whereas the other substructures were untouched. Since the substructures were much smaller than the global structure,the proposed method significantly improved the accuracy and efficiency of the model updating process.%提出了一种基于子结构的有限元模型修正方法。该方法将整体结构有限元模型划分为多个子结构模型，求解独立子结构的主模态特征解和特征灵敏度；通过位移协调条件和能量方程，约束相邻独立子结构，得到整体结构的特征解和特征灵敏度；并以整体结构模态和结构试验模态的残差为目标函数，通过调整子结构单元参数，完成有限元模型修正。当结构局部参数发生变化，通过分析某一个或几个子结构即可求解整体结构特征解和特征灵敏度，而不需要分析其他未发生变化的子结构。由于子结构模型尺寸远小于整体结构，该方法能够极大地提高有限元模型修正方法的精度和效率。