Research Article

An Improved Feature Parameter Extraction Algorithm of Composite Detection Method Based on the Fusion Theory

Table 1

Feature parameter description of PEC and UT response curve.

LabelParameterMeaningSource

1PeakThe maximum voltage over a period, except for fluctuations at the beginning and endPEC
2Peak timeThe time it takes to reach the peak (-axis)PEC
3Zero-crossing timeThe time difference of zero between the first and the second voltage valuesPEC
4Mean value PEC
5Mean square error PEC, UT
6Pulse index PEC
7Crooked degree index PEC
8Kurtosis index PEC, UT
9Amplitude of fundamental componentThe amplitude of the first peak of the response curve in the frequency domainPEC
10Amplitude of third harmonic componentThe amplitude of the third peak of the response curve in the frequency domainPEC
11Defect band meanThe mean ordinate of the bands with significant defect signals in the differential response curveUT
12Maximum differential peakThe maximum peak value of the wave caused by the difference between the defect and the nondefect signals in the differential signalUT
13Second differential peakIn the differential signal, the second peak value of the wave is caused by the difference between the defect and nondefect signals.UT