基于超声背散射信号处理的碳纤维复合材料孔隙检测研究

发布时间：2018-05-28 16:29

本文选题：碳纤维复合材料 + 超声波检测技术　；参考：《浙江大学》2016年博士论文

【摘要】：碳纤维复合材料(Carbon Fibre Reinforced Plastics,简称CFRP)在航空航天、车辆制造、大型工程建设等领域有着广泛应用。在制造和使用过程中,CFRP内部难免会出现缺陷,超声检测技术则是CFRP缺陷无损检测的主要手段之一。随着CFRP制造技术的进步,含有厚截面和曲面变厚度等区域的CFRP构件已经逐渐得到使用,这使得传统的复合材料超声检测手段及信号处理技术已很难再满足这些构件中某些部分的检测精度要求。针对这些材料的检测难点,本文以CFRP超声检测的相关基金项目为依托,对厚截面CFRP和曲面变厚度CFRP孔隙缺陷的超声检测技术进行研究,全文的研究工作及成果如下。(1)对各类CFRP孔隙超声检测方法进行分析,得到各方法的特点。同时采用金相实验法对论文涉及的CFRP的孔隙进行观察与分析,得到孔隙的形态与分布特征。根据上述分析结果提出了基于超声背散射信号处理的厚截面CFRP和曲面变厚度CFRP孔隙检测方法。(2)对声波在层状粘滞媒质中的反射与透射进行推导,得到声波反射系数分布函数频域模型。采用该模型对多层CFRP声波反射系数进行计算,得到超声波在层状CFRP中产生共振的条件及CFRP层数对共振的影响。进一步采用该频域模型对含孔隙层状CFRP声波反射系数进行计算,得到孔隙含量和分布对超声波共振的影响。同时,还对层状CFRP超声脉冲反射信号各成分特征进行分析,在此基础上建立了超声检测信号的时域及频域模型。(3)提出了基于超声背散射信号处理的厚截面CFRP局部集中孔隙缺陷识别方法。根据超声背散射信号特征将其划分为近表面信号和远表面信号。针对近表面信号提出了基于共振结构噪声特征与基于共振结构噪声去除这两种处理方法。针对远表面信号则主要提出了基于信号相关分析的小波变换模极大值去噪方法。采用上述方法对厚截面CFRP局部集中孔隙进行识别,通过破坏性金相实验验证了上述信号处理方法的可行性。(4)提出了基于超声背散射信号提升小波分解处理的曲面变厚度CFRP孔隙缺陷识别方法。通过金相实验测定了超声检测完毕的曲面变厚度CFRP试块孔隙率,同时分析了超声脉冲反射信号的特征。采用提升小波变换对超声背散射信号进行分解并分析得到了原始信号与各分解信号的特征。进一步对原始信号与选出的分解信号的特征随孔隙率的变化关系进行分析,结果表明最优分解信号特征比原始信号特征能更好地表征材料孔隙率。(5)提出了基于背散射信号能量特征的厚截面CFRP超声C扫描成像方法和基于背散射分解信号能量特征的曲面变厚度CFRP超声C扫描成像方法。同时,在基于第(3)点研究的基础上提出了厚截面CFRP超声背散射信号特征三维成像技术,生成的三维图像能够直观地对厚截面CFRP局部集中孔隙进行表征。
[Abstract]:Carbon Fibre Reinforced Plastics, (CFRP) is widely used in aerospace, vehicle manufacturing, large engineering construction and so on. In the process of manufacture and use, defects will inevitably appear in CFRP, and ultrasonic testing technology is one of the main methods for nondestructive testing of CFRP defects. With the development of CFRP manufacturing technology, CFRP components with thick cross-sections and curved surfaces with variable thickness have been gradually used. This makes it difficult for traditional ultrasonic testing methods and signal processing techniques of composite materials to meet the precision requirements of some parts of these components. In view of the difficulties of testing these materials, this paper studies the ultrasonic detection technology of thick section CFRP and curved surface variable thickness CFRP pore defect based on the related fund items of CFRP ultrasonic detection. The research work and results are as follows: 1) the ultrasonic testing methods of CFRP pore are analyzed, and the characteristics of each method are obtained. At the same time, the porosity of CFRP was observed and analyzed by metallographic experiment, and the pore morphology and distribution characteristics were obtained. Based on the above analysis results, a method of detecting thick section CFRP and curved surface variable thickness CFRP pore based on ultrasonic backscattering signal processing is proposed to deduce the reflection and transmission of acoustic waves in layered viscous media. The frequency domain model of acoustic reflection coefficient distribution function is obtained. By using this model, the reflection coefficients of multilayer CFRP sound waves are calculated, and the conditions of ultrasonic resonance in layered CFRP and the influence of the number of CFRP layers on the resonance are obtained. Furthermore, the frequency domain model is used to calculate the acoustic reflection coefficient of layered CFRP with pores, and the effects of pore content and distribution on ultrasonic resonance are obtained. At the same time, the characteristics of each component of the layered CFRP ultrasonic pulse reflection signal are analyzed. On this basis, the time-domain and frequency-domain models of ultrasonic detection signals are established. A method for identifying local concentrated pore defects in thick cross-section CFRP based on ultrasonic backscattering signal processing is proposed. According to the characteristics of ultrasonic backscattering signal, it is divided into near surface signal and far surface signal. In this paper, two processing methods for near-surface signals are proposed, which are based on resonance structural noise characteristics and resonance structural noise removal. For the far surface signal, a wavelet transform modulus maximum denoising method based on signal correlation analysis is proposed. The method is used to identify the local concentrated pores in thick section CFRP. The feasibility of the above signal processing method is verified by destructive metallographic experiments. (4) A surface variable thickness CFRP pore defect identification method based on ultrasonic backscattering signal lifting wavelet decomposition is proposed. The porosity of the curved surface CFRP specimen with varying thickness was measured by metallographic experiments and the characteristics of ultrasonic pulse reflection signal were analyzed at the same time. The lifting wavelet transform is used to decompose the ultrasonic backscattering signal and the characteristics of the original signal and the decomposed signal are obtained. Further, the relationship between the characteristics of the original signal and the selected decomposition signal with porosity is analyzed. The results show that the optimal decomposed signal feature can better characterize the material porosity than the original signal feature.) A thick cross-section CFRP ultrasonic C-scan imaging method based on backscatter signal energy characteristics and a backscatter decomposition signal energy are proposed. The method of curved surface variable thickness CFRP ultrasonic C scan imaging is presented in this paper. At the same time, based on the research of the third point, a 3D imaging technique of thick cross-section CFRP ultrasonic backscattering signal is proposed. The generated 3D image can directly characterize the local concentrated pores of the thick cross-section CFRP.
【学位授予单位】：浙江大学
【学位级别】：博士
【学位授予年份】：2016
【分类号】：TB33;TN911.7

【参考文献】