Identification of Acoustic Emission Signal of Tank Bottom Corrosion Based on Weighted Fuzzy Clustering
The identification of corrosion acoustic emission signal of tank bottom is the basis of improving detection accuracy. However, under some circumstances, the types of corrosion acoustic emission signal of the tank bottom are unknown, and a weighted fuzzy clustering recognition method was proposed to solve this problem. The main characteristics of the signal were given by acoustic emission detector; the characteristics included rise time, count, energy, duration, amplitude, and average frequency. Aiming at the randomness of clustering initialization, nearest neighbor method was used for optimizing initial clustering. For improving the accuracy of edge data, a weighted fuzzy clustering method was proposed to increase the difference of various acoustic emission signals. The data redistribution method of fuzzy clustering is adjusted with the weighted distance between the gravity and center to substitute the traditional distance and then allocated data to the set which had the minimum weighted distance. This study shows that the veracity of weighted fuzzy clustering increases by about 9 %.
KeywordsAcoustic Emission Cluster Center Storage Tank Fuzzy Cluster Acoustic Emission Signal
This work is supported by the Creative Team Project Foundation of the Education Department of Liaoning Province, China (Grant No. LT2010082).
- 4.W. Jin, C.Z. Chen, Z.H. Jin, B. Gong, B.C. Wen, The three-ratio method of acoustic emission source recognition [J]. Chin. J. Sci. Instrum. 29(3), 530 (2008)Google Scholar
- 5.G.Y. Gong, Study of FCM algorithms on parameters and its applications. Xi'an University of electronic science and technology [D] (2004)Google Scholar
- 8.X. Zhang, G. Zhang, P. Liu, Based on the clustering criterion function improved K-means algorithm. Comput. Eng. Appl. 47(11), 123 (2011)Google Scholar
- 9.S. Nittel, T.L. Kelvin, A. Braverman, Scaling clustering algorithms for massive data sets using data streams [C]. Proceedings of the 20th International Conference on Data Engineering, ICDE’04, 2004, p. 830Google Scholar
- 10.P.S. Bradley, U. Fayyad, C. Reina, Scaling clustering algorithms to large databases [C]. Proceedings of the 4th ACM SIGKDD, New York, 1998, pp. 9–15Google Scholar
- 12.W.K. Wang, Z.M. Zeng, G. Du, Y.J. Wei, S.Z. Song, Clustering analysis of acoustic emission signals during stress corrosion of 304 stainless steel [J]. CIESC Journal. 62(4), 1027 (2011)Google Scholar
- 13.W.K. Wang, Y.B. Li, G. Du, T. Zhang, S.J. Jin, Based on cluster analysis of acoustic emission signal fusion method at the tank bottom. J. Vibr. Shock. 31(17), 181 (2012)Google Scholar