Defect classification based on wavelet packet energy through pulsed alternating current field measurement technique

View Fulltext

Author(s): Jiuhao Ge¹ ; Chenkai Yang¹ ; Ping Wang¹ ; Yongsheng Shi²
- Affiliations: 1: Non-destructive Detection and Monitoring Technology for High-Speed Transportation Facilities , Key Laboratory of Ministry of Industry and Information Technology, Nanjing University of Aeronautics and Astronautics , Nanjing, Jiangsu, 210016 , People's Republic of China ;
  2: Railway Infrastructure Inspection Center, Chinese State Railway Group Co., Ltd. , Beijing 100081 , People's Republic of China
Source: Volume 56, Issue 19, 17 September 2020, p. 1016 – 1019
DOI: 10.1049/el.2020.1574 , Print ISSN 0013-5194, Online ISSN 1350-911X

In this Letter, a wavelet packet energy-based algorithm is proposed to classify surface and non-surface slits through the pulsed alternating current field measurement technique. Wavelet packet decomposition is used to obtain the coefficient of each frequency band. The slit is classified as non-surface slit or surface slit according to the normalised energy of the first band. The quantification of the remaining ligament or depth of slits is performed using the entropy of the wavelet packet energy. The experiment results show that the non-surface and surface slits can be clearly classified. The entropy of a non-surface slit is linearly proportional to the remaining ligament, and the entropy of a surface slit is inversely proportional to slit depth.

References

1. 1)
  - 2. Mirshekar-Syahkal, D., Mostafavi, R.F.: ‘Analysis technique for interaction of high-frequency rhombic inducer field with cracks in metals’, IEEE Trans. Magn., 1997, 33, (3), pp. 2291–2298 (doi: 10.1109/20.573845).
2. 2)
  - 4. Li, W., Chen, G., Yin, X., et al: ‘A feed-through ACFM probe with sensor array for pipe string cracks inspection’, NDT & E Int., 2014, 67, pp. 17–23 (doi: 10.1016/j.ndteint.2014.06.006).
3. 3)
  - 6. Papaelias, M.P., Lugg, M.C., Roberts, C., et al: ‘High-speed inspection of rails using ACFM techniques’, NDT & E Int., 2009, 42, (4), pp. 328–335 (doi: 10.1016/j.ndteint.2008.12.008).
4. 4)
  - 8. He, Y., Luo, F., Pan, M., et al: ‘Defect classification based on rectangular pulsed eddy current sensor in different directions’, Sens. Actuators A, Phys., 2010, 157, (1), pp. 26–31 (doi: 10.1016/j.sna.2009.11.012).
5. 5)
  - 5. Ge, J., Li, W., Chen, G., et al: ‘Analysis of signals for inclined crack detection through alternating current field measurement with a U-shaped probe’, Insight, Non-Destr. Test. Cond. Monit., 2017, 59, (3), pp. 121–128 (doi: 10.1784/insi.2017.59.3.121).
6. 6)
  - 9. Liu, Z., Cui, Y., Li, W.: ‘Combined power quality disturbances recognition using wavelet packet entropies and S-transform’, Entropy, 2015, 17, (8), pp. 5811–5828 (doi: 10.3390/e17085811).
7. 7)
  - 1. Lewis, A.M., Michael, D.H., Lugg, M.C., et al: ‘Thin-skin electromagnetic fields around surface-breaking cracks in metals’, J. Appl. Phys., 1988, 64, (8), pp. 3777–3784 (doi: 10.1063/1.341384).
8. 8)
  - 7. Mostafavi, R.F., Mirshekar-Syahkal, D.: ‘AC fields around short cracks in metals induced by rectangular coils’, IEEE Trans. Magn., 1999, 35, (3), pp. 2001–2006 (doi: 10.1109/20.764900).
9. 9)
  - 3. Li, W., Chen, G., Ge, J., et al: ‘High sensitivity rotating alternating current field measurement for arbitrary-angle underwater cracks’, NDT & E Int., 2016, 79, pp. 123–131 (doi: 10.1016/j.ndteint.2016.01.003).

Defect classification based on wavelet packet energy through pulsed alternating current field measurement technique

References

Related content