Research On Joint Scanning Laser Infrared Thermography Defect Detection Technology Of Carbon Fiber Reinforced Polymer Material

With the development of carbon fiber composite materials manufacturing and processing technology,due to its low density,high strength and high temperature resistance,carbon fiber composite materials have been widely used in aerospace,automotive lightweight,medical health,machinery automation and energy transportation.Carbon fiber composite materials are prone to damage and defects during service.The existing detection methods are difficult to meet the needs of noncontact,visualization,high resolution,large-area rapid detection and deep-buried defect detection,to ensure the safe and stable service of carbon fiber composite materials and promote its popularization and use.Starting from the research of carbon fiber composite material detection technology,this paper focuses on the joint scanning laser thermography detection technology(JSLT),and conducts theoretical and experimental analysis on the approximate model of JSLT,research on data reconstruction and spatiotemporal misalignment and spatial deformation correction methods and systematical research on the deconvolution thermal reconstruction method of the lateral thermal diffusion blur effect in the thermography detection system.The main research contents and innovations are as follows:1)The heat conduction model excited by the line shape joint scanning laser and the classic defect feature extraction method in the infrared thermography inspection system are studied.Based on the theoretical analysis of the data distribution in the x-y-t space the line shape joint scanning laser thermography,starting from the onedimensional heat conduction model of semi-infinite plane uniform excitation,the approximate model of the line shape joint scanning laser excitation is,the newly deduced model is similar with the surface pulse excitation model.Finally,from the perspective of data processing,thermographic signal reconstruction(TSR),principal component analysis(PCA),independent component analysis(ICA)and non-negative matrix factorization(NMF)are introduced,pointed out that these feature extraction methods meet the conditions of the line shape joint laser scanning excitation approximate model,it is feasible to combine these methods and joint scanning thermography detection system.2)The joint scanning laser thermography detection system is studied,and the data reconstruction and spatiotemporal misalignment and spatial deformation correction methods are developed to solve the problems in joint scanning thermography detection system.These methods break through the limitation that the needs to keep the scanning speed fully matched with the sampling frequency of the thermal imager to ensure that there is no deformation in traditional scanning experimental system.And these methods can simplify the parameter control of the experimental system.Reconstruction will increase the equivalent sampling frequency of the data and improve the time resolution of the infrared image sequence.The spatial deformation correction based on interpolation will improve the spatial resolution of the infrared image and increase the detection range.3)The experiment of JSLT inspection system for flat-bottom hole defect detection of carbon fiber composite material is studied.And it is verified that the traditional algorithms widely used in static infrared inspection system also perform well in JSLT system.Data processing results of different algorithms in JSLT and pulse thermography(PT)are compared and analyzed,under the assumption that defect diameter and buried depth have a linear relationship with the detection ability,when the defect diameter and buried depth are close,the JSLT can detect defects with a smaller diameter-todepth ratio than PT.The ratio has reached 3.3,which is more capable of detecting defects.At the same time,the JSLT experimental system can inspect a larger area of carbon fiber composite material in a shorter inspection time.4)The Lucy-Richardson deconvolution thermal reconstruction method in JSLT method based on is studied,which improves the detection ability of deep defects.Based on the approximate excitation model of JSLT,the thermal diffusion equation and the Lucy-Richardson nonlinear iterative method,a new lateral thermal diffusion deconvolution weakening method is developed.This method is applied to the detection of internal delamination defects in CFRP base material coating system.The results are compared with PCA,restored pseudo heat flux(RPHF),FFT and NMF methods.The experimental results show that the infrared image calculated by the method in this paper has a higher signal-to-noise ratio.The diameter-to-depth ratio of the defect detection can reach 1.5.The method performs better in the joint scanning laser thermal imaging system than other methods.At the same time,it has a certain suppression effect on the noise in the model approximation process and meets the requirements for rapid and effective detection of internal delamination defects in the carbon fiber composite base coating.Finally,a set of carbon fiber composite material inspection system that can realize non-contact,visualization,high-resolution,large-area rapid detection and deep-buried defect detection is established,which can provide a stable and reliable detection method for carbon fiber composite materials and coating systems.