Improved truncation singular value method for overcoming correction ill-condition of structural model

Abstract

The invention relates to the field of structural model correction, and specifically to an improved truncation singular value method for overcoming the correction ill-condition of a structural model. Finite element modeling is carried out on the structure through finite element software; a static test or a dynamic test is carried out on the structure; a static response parameter or a dynamic response parameter is obtained; a structure model correction basic equation is obtained; singular value decomposition is performed on a corrected basic equation coefficient matrix, correction parameter vectors, when only previous t singular values are taken, and corresponding residual vectors are calculated through a truncation singular value formula, corresponding scatter diagrams are drawn, all points are connected through straight line segments to obtain a broken line diagram, and a final model correction parameter solution can be obtained through the formula calculation. After the scheme is adopted, the method is suitable for solving the problem of finite element model correction of a large-scale structure, so that a stable and accurate model correction parameter solution can be obtained under the condition that data has errors.

Description

technical field [0001] The invention relates to the field of structural model correction, in particular to an improved truncated singular value method for overcoming ill-conditioned structural model correction. Background technique [0002] An accurate structural finite element model is the basis for structural mechanics analysis, response prediction, vibration control, damage identification, etc. Usually, the structural model established by the finite element software often has a large error with the real structure, which is reflected in the response parameters (such as static displacement, vibration frequency or mode shape) calculated by the structural finite element model and the actual structure of the real structure. The measured response parameters vary greatly. In this case, the structural model must be corrected to make the calculated response parameters as close as possible to the measured response parameters. The corrected finite element model can be used for struc...

AI Technical Summary

Problems solved by technology

[0010] The purpose of the present invention is to provide an improved truncated singular value method for overcoming the ill-conditionedness of structural model correction, which is suitable for solving the problem of finite element model correction of large structures, so that it can obtain stable and accurate results even when there are errors in the data. The model correction parameter solution of

Images