基于输入输出谱的模态参数识别方法

摘要

Owing to the FRF cannot be estimated exactly when only a limited test data is available,classical modal parameters identification algorithms based on FRF are not suit anymore.Modal identification algorithm based on input and output spectrums is especially suitable for this case.Common denominator model of FRF was adopted in the algorithm and the modal parameters were achieved by using least square estimator in this paper.The effects of leakage and initial condition could be eliminated by adding transient polynomial in system model.The bad numerical condition of polynomial coefficients could be avoided in discrete time domain.A simulation case of a 2-storey building and an application case of an automobile chassis were employed to validate the algorithm.The results showed that the method had a better identification result with a few data.%当测量数据有限时,准确估计频响函数比较困难,导致使用频响函数作为原始数据的模态参数识别方法受到限制.为此实现了一种使用输入、输出数据DFT谱的模态参数识别方法,无需估计频响函数.该方法中频响函数的参数模型采用公分母模型,使用线性化最小二乘法估计模态参数.在系统模型中添加瞬态项多项式减少了初始条件与泄漏的影响；在离散域内求解多项式系数,改善了数值解的病态问题.采用一个二层框架的仿真算例与汽车底盘实测算例对所给方法进行了验证,结果表明:该方法在仅使用少量数据的情况下能够取得较好的识别效果.