Research On Bonding Defect Detection Of New Lightweight Ceramic Matrix Composites Based On Ultrasonic Wave

The new ceramic matrix composites(CMC)are widely used in the thermal protection structure of spacecraft enclosures due to their excellent thermal insulation properties.In the process of sticking to the base of the aircraft,the thermal protection structure may cause bonding defects due to the craft problems.And the thermal damage caused by the atmosphere may also cause defects in the material during the service process.These damages can result in a decrease of the thermal insulation performance in the material and even make a lose of the thermal protection structure.The loss of the thermal protection structure seriously threaten the flight safety of the spacecraft and the personal safety of the astronauts.Therefore,the detection of bonding defects in new CMC materials is very important.In this paper,the ultrasonic C-scan method is used to detect the bonding defects of the material,and the research is carried out in three aspects: numerical simulation,signal processing and quantitative analysis.(1)In order to study the propagation characteristics of ultrasonic in new CMC materials,a numerical simulation method based on Gaussian acoustic beam model and finite element is adopted.Firstly,the theory of ultrasonic propagation characteristics in the material is analyzed and summarized.Based on this,the Gaussian sound beam model is used to simulate the sound field distribution of the ultrasonic transducer.Based on the finite element numerical calculation method,the transient propagation process of ultrasonic in the material is simulated.The numerical simulation technique adopted in this paper can provide theoretical basis and technical support for studying the ultrasonic propagation characteristics of CMC materials.(2)Based on the above theoretical analysis,this paper independently developed an ultrasonic C-scan automatic detection system,realizing the real-time display and pseudo3 D display of the detected images,thus completing the visual analysis of the ultrasonic detection data.The results show that the system is feasible for the detection of voids,debonding and inclusion defects in the CMC materials.(3)Aiming at the characteristics of multi-noise and signal analysis difficulty ofultrasonic data,a time-frequency analysis method combining HHT and pseudo Wigner-Ville distribution is proposed.Firstly,adaptive median filtering and wavelet threshold denoising algorithm are used for denoising,which greatly reduces the pulse and Gaussian noise carried by ultrasonic data.Then,based on the denoising,the time-frequency analysis of the detected signal is carried out by using HHT and pseudo Wigner-Ville distribution.The amplitude-frequency-time three-dimensional distribution of the signal is obtained,and the time-frequency domain feature analysis of the signal is realized.The results show that the time-frequency analysis method proposed in this paper can observe the distribution characteristics of signals in the time domain and frequency domain more intuitively and concretely.(4)In order to realize the quantitative and localization analysis of defects,the pulsed coupled neural network algorithm is used to segment the detected images after denoising,and the binary images segmented perform defect rounding and fitting processing,and compared with various segmentation methods.The results show that the quantitative and localization errors of the proposed method are the smallest,5% and 6% respectively.