Title :
Pipe defect sizing with matching pursuit based on modified dynamic differential evolution algorithm to recognize guided wave signal
Author :
Chuanjun, Shen ; Yuemin Wang ; Fangjun, Zhou ; Fengrui, Sun
Author_Institution :
Coll. of Archit. & Power, Naval Univ. of Eng., Wuhan, China
Abstract :
The modified dynamic differential evolution (MDDE) algorithm is proposed and applied to matching pursuit (MP) method. Two kinds of steel pipes with blind holes, and notches of different size, respectively, are tested by guided wave test system. The measured signals are decomposed by MP and the defect signals are easy to recognize from the processed signals. The average square root of the residual signals (ASRRS) after each matching pursuit iteration are calculated. The decline trend of ASRRS becomes greater when the defects are bigger. This feature can work as a criterion to estimate the defect size of steel pipes.
Keywords :
evolutionary computation; iterative methods; mechanical engineering computing; pattern matching; pipes; signal processing; steel; average square root of the residual signals; blind holes; defect signal recognition; defect size estimation; guided wave signal recognition; guided wave test system; matching pursuit iteration; measured signal decomposition; modified dynamic differential evolution algorithm; pipe defect sizing; steel pipes; Algorithm design and analysis; Approximation algorithms; Heuristic algorithms; Matching pursuit algorithms; Signal processing algorithms; Steel; Time frequency analysis; defect; differential evolution; guided wave; matching pursuit; recognize;
Conference_Titel :
Electronic Measurement & Instruments (ICEMI), 2011 10th International Conference on
Conference_Location :
Chengdu
Print_ISBN :
978-1-4244-8158-3
DOI :
10.1109/ICEMI.2011.6038007
