Nondestructive Evaluation Of Voids Content In Carbon Fiber Refinforced Plastics Based On Signal Processing

Carbon Fiber Reinforced Plastics(CFRP) was widely used in plane industry and many other fields, and the evaluation of voids content is necessary to ensure the safety of equipments. The morphology of voids in CFRP and techniques of Non-destructive evaluation of voids content were studied in this dissertation. Signal processing techniques such as Fourier Transformation, Wavelet Transformation and Lifting Schemes were used to process the ultrasonic signals obtained from experiments, and a portable voids content evaluating system was designed and fabricated.Firstly, the wide application of CFRP in various kinds of fields was introduced, then the common defects in CFRP, the importance of evaluation of voids contents and the existing evaluating methods were analyzed. Several signal processing methods and their application in Non-destructive testing were generalized and the significance of Non-destructive evaluation of voids content in CFRP was shown.The ultrasonic detecting theory in time domain and the factors affect the attenuation of ultrasonic were studied, and a detecting method in time domain was studied in chapter 2. Then the spectral analysis was introduced and the relationship between ultrasonic attenuation and frequency under various voids contents was studied. Furthermore, the change of the shape of spectral under various voids contents and the relationship between ultrasonic attenuation coefficients and voids content were studied. The formula of calculating phase velocity of ultrasonic in CFRP was obtained, and the phase velocity of ultrasonic under various voids content was studied. The phase velocity obtained by phase unwrapping was compared with those obtained by K-K relation, and the relationship between the change of phase velocity and voids content was studied.Morphology of voids in CFRP was studied in chapter 3. Microscope was used to study the properly treated surfaces of CFRP samples and the morphological characters of voids were analyzed. The design of the software and hardware of a portable voids content detecting system was introduced and a testing sample was given in this chapter.The basic theory of wavelet analysis and wavelet denoising were introduced in chapter 4, and three different series of wavelets (coifs, syms and dbs) were used to denoise the ultrasonic signals obtained from experiments. The Signal Noise Ration(SNR) and the Root Mean Square Error(RMSE) were analyzed using various wavelets and threshold methods, and the optimal wavelet and threshold method were obtained.The three steps of lifting scheme, Laurent Polynomial and Euclidean algorithm were studied in chapter 5, and several examples of construction of traditional wavelets were given.The steps of lifting schemes (splitting, prediction and update) were discussed in detail in chapter 6 and the calculating formula of several commonly used lifting schemes were presented. The denoising method using lifting scheme were analyzed and five different lifting schemes were used to denoise the ultrasonic signals.