Uniform design method for excitation parameter optimization of electromagnetic frequency mixing detection

Abstract

The invention discloses a uniform design method for excitation parameter optimization of electromagnetic frequency mixing detection. The method comprises the following steps: carrying out a uniform design test by taking four excitation parameters of frequency mixing detection as test factors, and obtaining frequency mixing detection signals under different excitation parameter combinations throughthe detection test; processing the frequency mixing detection signals, extracting various characteristic indexes capable of reflecting frequency mixing response from the frequency mixing detection signals, and constructing a fuzzy model for frequency mixing response evaluation based on the characteristic indexes; and establishing a multiple regression equation of the four excitation parameters according to the fuzzy evaluation result. According to the method, the influence of the excitation parameters on the electromagnetic frequency mixing response is analyzed by adopting a stepwise regression method, and the extreme value of regression equation is solved by utilizing a genetic algorithm, so that the optimal combination of the four excitation parameters is obtained.

Description

technical field [0001] The invention relates to a method for optimizing excitation parameters of electromagnetic frequency mixing detection, in particular to a multi-excitation parameter optimization method based on a uniform design test. The method is suitable for excitation parameter optimization of electromagnetic mixing detection of ferromagnetic materials, and belongs to the field of nondestructive testing. Background technique [0002] Electromagnetic mixing detection technology is a non-destructive testing method that uses electromagnetic nonlinear mixing effect to detect micro-damages in ferromagnetic materials. In electromagnetic mixing detection, ferromagnetic components need to be in the alternating magnetic field with high and low frequencies superimposed. Generally, the strength of the low-frequency magnetic field is relatively large, which can magnetize the component to a near-saturation state; the intensity of the high-frequency magnetic field is generally sm...

AI Technical Summary

Problems solved by technology

However, due to the uniform dispersion and orderly comparability of test points in the design of orthogonal experiments [4], the range and step size of the optimization parameters are limited, and it is not suitable for parameter optimization with more experimental levels.

Images