Research On Infrared Nondestructive Testing Technology For Water/Ice In Honeycomb Skin

The aircraft is prone to a variety of defects and damage during operation.Among them,structures such as the honeycomb of the aircraft fuselage can cause huge safety hazards due to water accumulation / icing.The infrared non-destructive testing method has considerable advantages in detecting internal defects in the structure.During the infrared image acquisition process,due to factors such as the quality of the equipment and the environment,a large amount of noise can be generated.The heat provided by the thermal excitation is uneven,resulting in uneven detection of defects.Clear and complete,increasing the difficulty of fuselage maintenance.Furthermore,the current infrared image processing still requires human intervention to complete the segmentation operation,and the accuracy and continuity of the segmentation results cannot meet the actual needs,and a method that can quickly and accurately detect and identify is needed to meet engineering applications.In this paper,combined with the infrared thermal image technology,the thermal excitation equipment such as flash lamp and hot-air gun is selected to heat the non-destructive specimen.Through the comparative analysis of the infrared images collected,it is found that the matrix U-shaped flash lamp as the thermal excitation equipment has the highest uniformity.In order to study the water / ice defect of honeycomb skin,referring to the Boeing 767 aviation nondestructive manual and other inspection methods about thermal imaging inspection of honeycomb structure parts,the test pieces of water / ice defect of honeycomb skin with different specifications of carbon fiber were made;meanwhile,aiming at the infrared image collected by the thermal imager,a variety of image denoising methods were analyzed and compared,and it was found that morphological filtering was the most effective in this experiment Method.Secondly,by comparing the detection results of many traditional edge detection operators and analyzing the defects of traditional edge detection operators,this paperproposes an improved Canny edge operator detection algorithm based on mathematical morphology filter and Otsu algorithm.Matlab simulation results show that the improved Canny edge algorithm can eliminate the interference of background noise to the target information,reduce the existence of false edge,and improve the correct detection rate CDR,false detection rate FDR,timeliness and other objective indicators to a certain extent.It is a practical and effective segmentation algorithm.Finally,based on the polynomial data fitting method of infrared sequence diagram,according to the different change rate of gray value(temperature)displayed by the thermophysical properties of different materials,this paper qualitatively determines the hydrops / ice accretion defect of honeycomb skin;at the same time,in order to carry out the quantitative identification of hydrops/ice accretion,the Canny algorithm and the region growing method are combined to automatically find the seed point of defect region and optimize the growth Criteria and stop growing conditions,improve the adaptability and segmentation effect,determine the defect area and defect lattice number.The final results show that the detection accuracy of this method is improved compared with the traditional segmentation method,and the adaptability of the algorithm is improved,which can provide reference for the follow-up maintenance scheme.