US 12,392,699 B1Granted

Defect detection method and device based on nonlinear system identification

AssigneeWuhan Institute of Technology

InventorsHanxin Chen, Shaoyi Li, Wenrui Yang, Kai Tan

Life of this patent

Estimating the term…

Loading patent…

Abstract

A defect detection method and device based on nonlinear system identification are provided. The defect detection method includes: performing, by a modal force hammer, hammer excitation on a test specimen according to a test schedule to generate an excitation signal and a response signal; acquiring, by a laser vibrometer, the excitation signal and the response signal; performing parameter calculation on the excitation signal and the response signal to obtain model parameters; constructing an initial Hammerstein model based on the model parameters; optimizing the initial Hammerstein model to obtain a Hammerstein model of the test specimen; adjusting the Hammerstein model of the test specimen by using a cross-validation method; perform defect determination based on the Hammerstein model of the test specimen adjusted and a pre-constructed template specimen model to obtain a determination result, and determining whether the test specimen is defective based on the determination result.

Claims9

Dependent claims

A defect detection method based on nonlinear system identification, comprising the following steps:

setting a test schedule according to a size of a test specimen;
performing, by a modal force hammer, hammer excitation on the test specimen according to the test schedule to generate an excitation signal;
and making the test specimen vibrate based on the excitation signal to generate a response signal;
acquiring, by a laser vibrometer, the excitation signal and the response signal;
performing nonlinear parameter calculation on the excitation signal and the response signal to obtain an initial nonlinear parameter;
performing linear parameter calculation on the excitation signal and the response signal to obtain an initial linear parameter;
constructing an initial Hammerstein model based on the initial nonlinear parameter and the initial linear parameter;
and optimizing the initial Hammerstein model to obtain a Hammerstein model of the test specimen;
and
adjusting the Hammerstein model of the test specimen by using a cross-validation method to obtain an adjusted Hammerstein model of the test specimen;
performing defect determination based on the adjusted Hammerstein model of the test specimen and a pre-constructed template specimen model to obtain a determination result;
and determining whether the test specimen is defective based on the determination result;
wherein the performing defect determination based on the adjusted Hammerstein model of the test specimen and the pre-constructed template specimen model to obtain the determination result, and determining whether the test specimen is defective based on the determination result comprises:
performing feature comparison on the adjusted Hammerstein model of the test specimen and the pre-constructed template specimen model to obtain comparison features;
and
determining whether the test specimen is defective based on the comparison features and preset defect conditions.

13 more elements

A defect detection device based on nonlinear system identification, comprising:

a signal generation unit, a signal acquisition unit, a model construction unit, and a defect detection unit;
wherein the signal generation unit is configured to:
set a test schedule according to a size of a test specimen;
perform, by a modal force hammer, hammer excitation on the test specimen according to the test schedule to generate an excitation signal;
and make the test specimen vibrate based on the excitation signal to generate a response signal;
wherein the signal acquisition unit is configured to acquire, by a laser vibrometer, the excitation signal and the response signal;
wherein the model construction unit is configured to:
perform nonlinear parameter calculation on the excitation signal and the response signal to obtain an initial nonlinear parameter;
perform linear parameter calculation on the excitation signal and the response signal to obtain an initial linear parameter;
construct an initial Hammerstein model based on the initial nonlinear parameter and the initial linear parameter;
and optimize the initial Hammerstein model to obtain a Hammerstein model of the test specimen;
and
wherein the defect detection unit is configured to:
adjust the Hammerstein model of the test specimen by using a cross-validation method to obtain an adjusted Hammerstein model of the test specimen;
perform defect determination based on the adjusted Hammerstein model of the test specimen and a pre-constructed template specimen model to obtain a determination result;
and determine whether the test specimen is defective based on the determination result;
and
wherein, in the defect detection unit, steps of performing the defect determination based on the adjusted Hammerstein model of the test specimen and the pre-constructed template specimen model to obtain the determination result, and determining whether the test specimen is defective based on the determination result comprise:
performing feature comparison on the adjusted Hammerstein model of the test specimen and the pre-constructed template specimen model to obtain comparison features, and determining whether the test specimen is defective based on the comparison features and preset defect conditions.

16 more elements

Dependencies are inferred from the claim text.