实验模态分析快速计算方法与应用研究

摘要

A method combining the model order reduction and its parameters optimization was proposed to achieve the rapid calculation for experimental modal analysis.The indirect model reduction method was used in least square complex frequency domain to achieve the order reduction of system characteristic matrix.An optimization model was established based on the parametric description of the characteristic matrix.The system's degrees of freedom were optimized and reduced using the differential integrated comprehensive optimization strategy.The genetic algorithm was used to obtain the global optimization and reduction of parameters.The application of the fast calculation method in the experimental modal analysis of a heavy truck cab's body-in-white proves the effectiveness of the method and the improvement in computational efficiency.The method establishes a solid foundation for online detection and diagnosis of structural dynamics characteristics under continuous excitation condition.%提出联合模型降阶和智能参数优化的方法实现实验模态分析的快速计算。在最小二乘复频域法中采用间接模型降阶法实现系统特征矩阵的降阶。将决定参数识别效率的特征矩阵进行参数化描述并应用遗传算法对优化模型参数进行全局优化和缩减。采用差异化综合策略重点对自由度进行分步优化和缩减。对重卡驾驶室白车身进行实验模态分析快速计算方法的应用并证明新方法对计算效率提高的有效性。该方法为实现连续不间断激励下系统结构动力学特征在线检测和诊断打下了坚实基础。