Affiliation of Author(s):能源与动力学院
Journal:Zhendong Gongcheng Xuebao
Abstract:An improved parallelepiped convex model is proposed to describe the wide correlation between uncertain parameters. In combination with response surface method, a dynamic uncertain parameter identification method based on non-probabilistic convex model is presented. First of all, the improved parallelepiped two-dimensional model is used to describe the correlation distribution of the dynamic response frequencies of the structure and then to construct the multidimensional model, which is transformed into the interval form of standard parameter space by affine coordinate transformation. Then, taking the response frequency in the standard parameter space as the input and the structure uncertain parameter as the output, we construct the complete-quadratic-term interval inverse response surface model, which can avoid the interval expansion problem caused by interval operation and directly carry out the interval operation to identify the interval of dynamic uncertain parameters. Finally, this method is verified by a simulation example of a 3DOF mass spring system and a modal experiment of a bolt connection frame. The results show that the proposed method can effectively identify the dynamic uncertain parameters of the structure with a high accuracy. © 2019, Nanjing Univ. of Aeronautics an Astronautics. All right reserved.
ISSN No.:1004-4523
Translation or Not:no
Date of Publication:2019-02-01
Co-author:Wang, Pan
Correspondence Author:Zang Chaoping