Detection And Characterization Of Adhesive Defect For Thermal Pretection Composite Based On Coplanar Electrical Capacitance Imaging

Hypersonic aircraft is a cutting-edge technology in aerospace,which has attracted many aerospace power countries to carry out research.Due to low adhesion quality of thermal insulating materials on the aircraft's substrate surface,it can easily cause the break of thermal insulating structure during flight task.Therefore,it is very important to detect the adhesive defects/damage and evaluate the adhesion quality of the thermal insulating structure,which is a favorable guarantee to realize the safe and stable operation of the aircraft.For testing of the structure adhesive quality of the thermal insulating composites,there is no perfect testing standards and evaluation system at present.Taking into account the demand of the high reliability and long life for the advanced thermal insulating composite structure,the testing principle and characterization techniques of the adhesion defects based on planar array capacitive imaging are focused researched.(1)The adhesive layer is between the aircraft substrate and the thermal insulating material,which makes access to bonding interface limited.This thesis presents a visualization method that can non-contact detect adhesive defect from the outside surface.Based on the sensing mechanism of planar array electrodes,the information of defects is reflected through the reconstructed image of dielectric constant distribution.(2)For a new type of multi-functional porous ceramic based fiber lightweight composite(the most advanced insulation tile),based on energy spectrum and nuclear magnetic resonance imaging,the microstructure and mechanical properties of the composite were characterized.By considering the insulation tile and adhesion layer as the measuring field,a theoretical analysis on sensitive characteristics of dielectric constant for planar array electrode is established.On the basis of sensitivity map,the relationship between the capacitance change and the defect feature is investigated.The optimization design of the 3x4 array electrode sensor is simulated and the performance optimization weighted function is proposed.The sensitivity matrix of the edge field is calculated by finite element simulation and its distribution characteristics are analyzed.The results of the experiment verified the sensing ability of the array electrodes sensor to a small defect with a minimum diameter less of 1mm.(3)In order to establish correlation between dielectric constant distribution and image gray level,planar array electrode capacitance imaging algorithm is studied to obtain characterization of the measured defects.The setting method of absolute empty field and full field with single medium leads to the problem of low resolution imaging.To improve image quality,a relative empty field and full field setting method by using multiple media is proposed.An improved Landweber iterative reconstruction algorithm is proposed to solve the problem of initial value deviation of iteration.An Kalman filtering algorithm is proposed to denoise the measured data,so as to improve the clarity of the reconstructed image.An improved PSO reconstruction algorithm is used to solve the problem that the iterative imaging algorithm relies on the initial iterative value.By using a capacitive imaging system based on planar 3x4 array electrode,the experimental research on bonding defect detection of porous thermal insulating material has been carried out.The experimental results show that the method presented can be used for imaging identification of defects with the identifiable minimum area of 5mm x5 mm.(4)The capacitance data redundancy according to contribution rate is analyzed.The unredundancy data is used to reconstruct and improve the quality of the image.The high dispersion of multi-layered reconstructed image results in the difficulty in judging the accurate information of defects.Aim at this problem,The data fusion method of principal component analysis is adopted.The final fusion image is obtained to show the detected defects.The block fractal dimension is calculated to position defects automatically,and the automatic localization error is less than 0.8mm.The traditional threshold segmentation technology is adopted to rebuild image segmentation area of defects,and the defect area can be figured out.(5)Through the test experiments of thermal insulating composite structure for hypersonic vehicle,the feasibility verification test for the planar array capacitance image sensing is conducted,which provides reliable technical support for the practical engineering application and a solid foundation for further improvement of health monitoring and assessment for the thermal insulating structure.