Traditional ultrasonic TOFD （ time of flight diffraction） has the major shortcoming of low amplitude of diffractive wave which brings about lack of sensitivity for weld defect detection. Aimed at the technological l...Traditional ultrasonic TOFD （ time of flight diffraction） has the major shortcoming of low amplitude of diffractive wave which brings about lack of sensitivity for weld defect detection. Aimed at the technological limitation, a novel TOFD method is proposed by developing a focusing probe. Through the analyses and calculation of sound field distribution based on geometric acoustics, a cylindrical surface wedge is designed and produced. Artificial defect containing testing piece is made and tested using both traditional and focusing TOFD, and the received signal and image are compared. The result shows that the proposed focusing method can converge the emitted sound energy effectively and improve testing sensitivity greatly. Compared with traditional TOFD tested data, focusing TOFD tested defect wave in A-scan line and defect diffractive stripe in D-scan image can be identified easily.展开更多
为解决超声渡越衍射时差(Time of flight diffraction,TOFD)检测图像中缺陷识别的问题,分析检测图像的特点,研究图像自动分割的算法,其中包括图像预处理及图像分割。提出基于信号互相关算法的图像预处理方法,在此基础上,应用峰值搜索方...为解决超声渡越衍射时差(Time of flight diffraction,TOFD)检测图像中缺陷识别的问题,分析检测图像的特点,研究图像自动分割的算法,其中包括图像预处理及图像分割。提出基于信号互相关算法的图像预处理方法,在此基础上,应用峰值搜索方法提取焊缝区域,利用带控制标记符的分水岭变换对预处理的图像进行分割,从而识别出缺陷目标。利用提出的图像自动分割方法分割不同的超声TOFD检测图像。研究表明,基于信号互相关算法的图像校正方法在一定程度上可抑制检测图像的畸变,焊缝区域图像的提取可减少图像分割过程的计算量,从局部极值的角度出发的带控制标记符的分水岭变换实现缺陷目标的分割。与基于阈值方法的图像分割结果相比,图像自动分割算法较好地解决了近表面缺陷的识别问题,同时提出的方法也可用于含多个缺陷的图像分割。展开更多
通过Hamming窗加权方法设计了幅度加权调频编码激励信号,将这种新型的激励方法与TOFD(Time-of-flight diffraction)检测方法相结合,综合提高了粗晶奥氏体不锈钢焊缝缺陷检测的时间分辨力、检测信噪比和缺陷定量定位精度,有效改善了粗晶...通过Hamming窗加权方法设计了幅度加权调频编码激励信号,将这种新型的激励方法与TOFD(Time-of-flight diffraction)检测方法相结合,综合提高了粗晶奥氏体不锈钢焊缝缺陷检测的时间分辨力、检测信噪比和缺陷定量定位精度,有效改善了粗晶焊缝超声检测中的难点问题。为分析设计的幅度加权调频编码激励信号的检测能力,针对奥氏体不锈钢母材试件和焊缝试件中的横孔缺陷,采用5 MHz探头分别进行了编码激励和常规激励的TOFD成像检测对比试验,结果表明:在相同的检测条件下,幅度加权调频编码激励可提高了图像和信号质量,使检测信号中的杂波和噪声得到抑制,缺陷上端和下端衍射波被准确区分,使各波形的时间宽度降低了30%,有效提高了TOFD检测的时间分辨力;获得的缺陷定位定量测量的平均相对误差为3.8%,较常规激励降低了47%,这种激励方法可在不提高激励电压和增益条件下,使不锈钢焊缝中缺陷检测的信噪比达16 d B以上,较常规激励平均提高了7 d B。展开更多
基金Supported by the International Cooperation Project （2007DFR70070）, the National Natural Science Foundation of China （51005056, 50775054） and the Research Fund for the Doctoral Program of Higher Education （20102302120045 ）.
文摘Traditional ultrasonic TOFD （ time of flight diffraction） has the major shortcoming of low amplitude of diffractive wave which brings about lack of sensitivity for weld defect detection. Aimed at the technological limitation, a novel TOFD method is proposed by developing a focusing probe. Through the analyses and calculation of sound field distribution based on geometric acoustics, a cylindrical surface wedge is designed and produced. Artificial defect containing testing piece is made and tested using both traditional and focusing TOFD, and the received signal and image are compared. The result shows that the proposed focusing method can converge the emitted sound energy effectively and improve testing sensitivity greatly. Compared with traditional TOFD tested data, focusing TOFD tested defect wave in A-scan line and defect diffractive stripe in D-scan image can be identified easily.
文摘为解决超声渡越衍射时差(Time of flight diffraction,TOFD)检测图像中缺陷识别的问题,分析检测图像的特点,研究图像自动分割的算法,其中包括图像预处理及图像分割。提出基于信号互相关算法的图像预处理方法,在此基础上,应用峰值搜索方法提取焊缝区域,利用带控制标记符的分水岭变换对预处理的图像进行分割,从而识别出缺陷目标。利用提出的图像自动分割方法分割不同的超声TOFD检测图像。研究表明,基于信号互相关算法的图像校正方法在一定程度上可抑制检测图像的畸变,焊缝区域图像的提取可减少图像分割过程的计算量,从局部极值的角度出发的带控制标记符的分水岭变换实现缺陷目标的分割。与基于阈值方法的图像分割结果相比,图像自动分割算法较好地解决了近表面缺陷的识别问题,同时提出的方法也可用于含多个缺陷的图像分割。
文摘通过Hamming窗加权方法设计了幅度加权调频编码激励信号,将这种新型的激励方法与TOFD(Time-of-flight diffraction)检测方法相结合,综合提高了粗晶奥氏体不锈钢焊缝缺陷检测的时间分辨力、检测信噪比和缺陷定量定位精度,有效改善了粗晶焊缝超声检测中的难点问题。为分析设计的幅度加权调频编码激励信号的检测能力,针对奥氏体不锈钢母材试件和焊缝试件中的横孔缺陷,采用5 MHz探头分别进行了编码激励和常规激励的TOFD成像检测对比试验,结果表明:在相同的检测条件下,幅度加权调频编码激励可提高了图像和信号质量,使检测信号中的杂波和噪声得到抑制,缺陷上端和下端衍射波被准确区分,使各波形的时间宽度降低了30%,有效提高了TOFD检测的时间分辨力;获得的缺陷定位定量测量的平均相对误差为3.8%,较常规激励降低了47%,这种激励方法可在不提高激励电压和增益条件下,使不锈钢焊缝中缺陷检测的信噪比达16 d B以上,较常规激励平均提高了7 d B。
