@article{ref1,
title="Pipeline defect recognition method based on CEEMD-FCM",
journal="China safety science journal (CSSJ)",
year="2020",
author="Wang, C. and Liang, W. and Liang, X.",
volume="30",
number="1",
pages="87-93",
abstract="In order to improve identification accuracy of pipeline defects, CEEMD-FCM model is proposed by using CEEMD and FCM clustering algorithm. Firstly, based on analysis of defect signal waveform features, particle swarm optimization algorithm (PSO) was introduced to improve wavelet threshold de-noising method, and noise reduction of pipeline defect signals was realized. Then, defect signals were decomposed by CEEMD, and characteristic parameters of defects were extracted through principle of energy entropy. Finally, FCM was optimized by simulating annealing algorithm (SA) and genetic algorithm (GA) to complete classification of pipeline defects. The results show that comprehensive identification accuracy of the proposed identification method reaches 87. 5%. It can achieve accurate identification of defect modes in petrochemical industry, therefore ensuing safe operation of pipeline and reducing accident rates. © 2019 China Safety Science Journal<p /><p>Language: zh</p>",
language="zh",
issn="1003-3033",
doi="10.16265/j.cnki.issn1003-3033.2020.01.014",
url="http://dx.doi.org/10.16265/j.cnki.issn1003-3033.2020.01.014"
}