复杂异型机匣高效高精度动力学建模及评价方法

为了解决复杂异型机匣模型单元数量大、原始矩阵阶数高导致的动力学计算与后处理困难的问题，提出基于试验模态分析-大规模有限元-子结构缩聚的复杂异型机匣高精度动力学建模及评价方法。以某型直升机主减速器机匣为研究对象，建立该异型构件原始有限元模型并通过模态试验验证模型的有效性，通过分析机匣子结构各阶模态保留主振型，选择模态能量较大处为缩聚点，得到自由度数目大幅减少的缩聚模型，对比验证缩聚前后模态的一致性，并提出一种基于数列相关系数定义的缩聚误差衡量方法，最后利用界面位移协调条件进行子结构耦合，对比整体模型的固有特性以及计算效率。研究结果表明:缩聚矩阵与有限元原始矩阵动力学特性十分接近，固有频率与振型误差均小于4%，且计算时间更短，存储空间占用更少，极大地提高了计算效率。 

High-efficiency and accurate dynamical modeling and evaluation method for complex irregular casing

In order to solve the problem of difficult dynamic calculation and post-processing caused by complex irregular casing of large number of model elements and high order of the original matrix, a high-precision dynamic modeling and evaluation method for complex irregular casing based on experimental modal analysis-large-scale finite element-substructure condensation was proposed. Taking the main casing of a helicopter as the research object, the original finite element model of the special-shaped component was established and the validity of the model was verified through modal experiments. By analyzing the retained models of each mode of the substructure, the larger vibration energy space was selected as the condensed node, and then a condensation model with a greatly reduced number of degrees of freedom was obtained. The consistency of each mode before and after the condensation was verified by comparison. A method for measuring the model condensation error based on the correlation coefficient of the sequence was proposed. Finally, by taking the interface displacement compatibility condition into consideration, the substructure coupling was carried out; in addition, the inherent characteristics and computational efficiency of the overall model were compared. The research results showed that the dynamic characteristics of the condensation matrix were very close to the original matrix of the finite element, both the natural frequency and the mode error were less than 4%, the calculation time was shorter and the storage space was less occupied, greatly improving the calculation efficiency.