分离谱技术在超声衍射时差法信号处理中的应用

Application on split-spectrum technique in ultrasonic TOFD signal processing

超声衍射时差(TOFD)方法因具有普通超声检测和射线检测的优点而被广泛应用于中厚板的焊缝检测与缺陷定量中。TOFD技术检测的是相对微弱、指向性差的衍射波信号,被检测材料所产生的结构噪声也会降低检测信号的信噪比,影响了TOFD检测的精度。将分离谱技术用于超声TOFD检测信号的处理,采用线性平均、极值阈值、极值阈值+最小值、最小值等四种恢复算法进行比较,并在最小值算法的基础上引入最小值选中次数加权算法恢复信号。结果表明:与传统的滤波方法相比,该方法能有效地提高了回波信号的信噪比,减小了TOFD检测中的缺陷定量误差。

With the advantage of precise quantification, ultrasonic time-of-flight-diffraction （TOFD） technique has been widely using in weld detection and defect quantification of middle thick plate. The signals detected by using TOFD technique are relatively weak diffraction wave signals with poor directivity, and the structure noise generated by the detected material will also reduce the signal-to-noise ratio and therefore affect the detection sensitivity of TDFD technique.. The split-spectrum technique is used for processing the ultrasonic TOFD detecting signal in this paper. Four restoration algorithms, i.e. the linear average, the maximal threshold, the maximal threshold＋minimal value and the minimal value, are adopted for comparison. In the end the weighted algorithm of minimal number selected based on the minimal value algorithm is used to restore the signal. Compared with the traditional filtering method, the results show that this method can effectively increase the signal-to-noise ratio of echo signal and reduce the defect quantitative error of TOFD detection.