基于归一化频响函数曲率的随机模型修正

Stochastic Model Updating Based on Curvature of Normalized Frequency Response Function

为提高随机模型修正的计算效率以及待修正参数均值和标准差的修正精度,提出了一种将Kriging模型和归一化频响函数曲率相结合的随机模型修正方法。首先,构造Kriging模型,将待修正参数作为Kriging模型输入,对应的归一化频响函数曲率作为Kriging模型输出;然后,采用麻雀算法以Kriging模型预测响应与实验响应之差的绝对值之和作为目标函数修正参数均值,以互信息作为目标函数修正参数标准差;最后,通过三维桁架验证了所提方法的有效性。结果表明,所提方法修正后的参数均值误差均低于0.2%,标准差误差均低于2.5%。

In order to improve the computational efficiency and the precision of the mean and standard deviation of each parameter to be updated in the stochastic model updating,a stochastic model updating method combining Kriging model and normalized frequency response function curvature is proposed.Firstly,a Kriging model is constructed;the model parameters to be updated are as the Kriging model import;the corresponding normalized frequency response function curvature is as the Kriging model output;Then,the sparrow search algorithm is used to update the mean of each parameter by using the objective function of the sum of the absolute of the difference between the predicted response of Kriging model and the tested response,by taking mutual information as the objective function to update the standard deviation of each parameter;Finally,the effectiveness of the proposed method is verified by a three-dimensional truss.The results show that the means of the updated parameters error are less than 0.2%,and the errors of the standard deviations of the updated parameters are less than 2.5%.