| Carbon fiber reinforced silicon carbide ceramic matrix composite is a promising high temperature structural material.In order to ensure the quality of products,non-destructive testing is needed.Practice shows that the ultrasonic testing technology is suitable for the material.Yet the material in the inspection can not use the traditional liquid coupling agent,therefore air-coupled ultrasonic testing technique(referred to as ACUT)is the only suitable method.In view of the particularity of this material,a systematic study of the air coupled ultrasonic testing technology for this material is scarce at home and abroad.In this paper,the detection technology of delamination and porosity in C/SiC composites is studied.The main contents and results are as follows:(1)The sound field of the air coupled ultrasonic probe is studiedby the multi-gauss acoustic beam model and then corrected according to the air attenuation factor.Secondly,the sound field measurement system is designed tomeasure sound field of four kinds of probe,and the influence of the sound field on the detection effect of the delamination defect is analyzed.The results show that higher detection precision can be obtained by selecting high frequency focusing probe under the condition of penetration ability.And the sample should be placed in the focus position of the sound beam.Finally,the diffraction field of the edge of the sample is visualized by the system to analyze the generation mechanism of the blind area,and the corresponding precautionarymeasures are given.(2)The contact ultrasonic method shows that C/SiC composites has a sound velocity of about 2134m/s and a sound attenuationcoefficient of about 12.45dB/mm.The influence of tilt,frequency thickness,stratification and porosity on ACUT signal ofC/SiC composites is analyzed toprovide basis for the later signal analysis.Besides,the influence of the parameters of the detection system on the signal is investigation.The detection ability can satisfy the detection of defects about 8~10mm diameter.A 7mm thick C/SiC composite specimen was tested by air coupled ultrasonic and the test data were saved.Finally,the application of LFM pulse compression technology in detection is studied.(3)After radiographic CT test confirmed that there was natural lamination defect in the specimen,metallographic test are applied to this sample toestablishment the relationship between signal characteristics and delamination and 3 kinds of pore level in material.The causes of odd frequency multiplication and the method of improving spectrum accuracy are investigated.The time-frequency analysis is implemented by wavelettransform(WT)and Hilbert-Huang transform(HHT).The HHT analysis algorithm of C/SiC signal is optimized,and 3 step preprocessing is put forward to improve the effect of HHT time frequency graph.(4)The classification process of material status is put forward to divide the material state into 4 grades.Among them,two states of minimum porosity and common porosity are distinguished by neural network technology,which adopting three types features of signal——IMFs,HHT and time domain frequency domain combined eigenvalue as input eigenvalue,and the correct rate of recognition is all above 90%.Using MATLAB to design a software which can make analysis of ACUT data from C/SiC composites,the core features include: gate reset C scan imaging,feature imaging,neural network training,material state recognition based on neural network technology,and so on. |
PDF Full Text Request