Research On Optically Excited Thermography Nondestructive Detection Method Based On Sparse Decomposition Of Hilbert Time-Frequency Information

Composite materials are used in many fields due to their good properties,and defects such as debonding and delamination will occur in the process of preparation and use,causing safety hazards.Non-destructive testing and evaluation technology is an important means of judging defects in composite materials.Because of its high detection efficiency and intuitive results,optical excitation infrared thermography technology has been widely used in nondestructive testing of composite materials.In this paper,based on the nondestructive testing technology of photoexcitation infrared thermography,the use of composite materials and the status of defect detection are described,the infrared radiation theory is described,as well as the basic formula and model of heat conduction,and the one-dimensional heat conduction model is taken as the basis of analysis.This article describes the basic composition of the optically excited infrared thermal imaging system and the construction of the experimental platform,expounds the classic infrared thermal imaging data processing algorithm,and proposes the Hilbert transform time-frequency infrared signal processing method,which can combine the time domain and frequency domain at the same time Analyze and fully mine defect information.Based on the one-dimensional heat conduction formula combined with the Hilbert transform,this paper first derives the time-frequency formula of the defect area and the non-defect area,and proposes the basis for frequency imaging.On this basis,the positive correlation between defect appearance time and defect depth is deeply analyzed,and the feasibility of the method is verified through simulation analysis.Finally,test pieces with different types of different depth defects are selected for experimental verification.The main innovations of this article are as follows:(1)It is proposed to combine the Hilbert transform with the heat conduction formula and analyze the infrared characteristic information in the time-frequency domain at the same time to overcome the limitations of single-domain analysis.(2)Through theoretical analysis,the basis of Hilbert transform frequency imaging is proposed,combined with L4 norm sparse dictionary learning to detect defects,and the imaging signal-to-noise ratio is improved.The method is compared and verified with traditional independent component analysis.(3)Analyzed the positive correlation between the time of occurrence of defects and the depth of defects in Hilbert frequency imaging within a specific time range.