不确定性参数识别的模态区间逆响应面法

Method for Uncertain Parameter Identification Using the Modal Interval Inverse Response Surface

结合模态区间分析及响应面的相关理论,提出一种新的不确定性参数识别方法,即模态区间逆响应面法。首先,以有界区间数来量化结构参数的不确定性,通过合理的实验设计确定样本数据;然后,以响应为输入,设计参数为输出,采用逐步回归分析构造设计参数与结构响应的模态区间逆响应面模型,进而直接在模态区间逆响应面模型上进行模态区间运算,即可识别材料参数的变异性区间;最后,采用一组钢板模态实验来验证所提方法的可行性及可靠性。结果表明:所提方法可准确识别钢板材料参数的取值区间,有效地解决多重变量区间运算存在的区间过估计问题,识别过程避免区间迭代优化,具有较高的计算效率。

Combining the relevant theories of modal interval analysis with response surface, a new method of structural parameter identification with uncertainty, namely modal interval inverse response surface, was proposed. First, the interval variables were adopted to quantify the uncertainty of the model parameters. Then, through an experimental design to determine the design points, the mathematical model of responses and uncertain parameters was constructed based on the stepwise regression method, in which the responses of a structure were denoted as inputs and the design parameters as outputs. Finally, the modal interval analysis was applied to uncertainty parameter identification without the iterative optimization process. The feasibility and accuracy of the proposed method was verified using a set of plate modal tests. The results showed that the proposed method can improve calculation efficiency, accuracy, and the overall estimation of multivariable interval extension. © 2016, Editorial Department of JVMD. All right reserved.