Classification of Steam Generator Tube Defects for Real-Time Applications Using Eddy Current Test Data and Self-Organizing Maps

Mesquita, Roberto Navarro de; Ting, Daniel Kao Sun; Cabral, Eduardo Lobo Lustosa; Upadhyaya, B.R.

doi:10.1023/b:time.0000019126.26053.23

Cited by 6 publications

(4 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Classification algorithms for eddy current testing can be arranged into two main categories: Signal processing based: Eddy current signals are processed in order to extract some characteristics from them, allowing differentiating among defect types, as in [ 6 – 11 ]. Shape based: The shape of the eddy current signals in the impedance plane are processed to find out the contours or appearances associated with every type of defect, as in [ 12 – 15 ].…”

Section: Description Of the Problemmentioning

confidence: 99%

“…Signal processing based: Eddy current signals are processed in order to extract some characteristics from them, allowing differentiating among defect types, as in [ 6 – 11 ].…”

Section: Description Of the Problemmentioning

confidence: 99%

“…The Hilbert transform [ 9 ] and the principal component analysis [ 10 ] are some of the techniques used to extract features from the eddy current signals. Furthermore, neural networks have been applied in [ 11 ].…”

Section: Description Of the Problemmentioning

confidence: 99%

See 2 more Smart Citations

Unsupervised Classification of Surface Defects in Wire Rod Production Obtained by Eddy Current Sensors

Saludes-Rodil

Baeyens

Rodríguez-Juan³

2015

Sensors

View full text Add to dashboard Cite

An unsupervised approach to classify surface defects in wire rod manufacturing is developed in this paper. The defects are extracted from an eddy current signal and classified using a clustering technique that uses the dynamic time warping distance as the dissimilarity measure. The new approach has been successfully tested using industrial data. It is shown that it outperforms other classification alternatives, such as the modified Fourier descriptors.

show abstract

Section: Description Of the Problemmentioning

confidence: 99%

“…Signal processing based: Eddy current signals are processed in order to extract some characteristics from them, allowing differentiating among defect types, as in [ 6 – 11 ].…”

Section: Description Of the Problemmentioning

confidence: 99%

See 1 more Smart Citation

Unsupervised Classification of Surface Defects in Wire Rod Production Obtained by Eddy Current Sensors

Saludes-Rodil

Baeyens

Rodríguez-Juan³

2015

Sensors

View full text Add to dashboard Cite

show abstract

“…In addition, depending on the power of the reactor, two, three or four steam generators are provided and, together with the reactor, are installed in the hermetically sealed doubled-walled reactor building. The efficiency of a steam boiler depends on the heat transfer properties from the primary to the secondary fluid [39].…”

Section: Illustrative Examplesmentioning

confidence: 99%

A combination of genetic algorithm‐based fuzzy C‐means with a convex hull‐based regression for real‐time fuzzy switching regression analysis: application to industrial intelligent data analysis

Ramli

Watada

Pedrycz

2013

IEEJ Transactions Elec Engng

View full text Add to dashboard Cite

Processing an increasing volume of data, especially in industrial and manufacturing domains, calls for advanced tools of data analysis. Knowledge discovery is a process of analyzing data from different perspectives and summarizing the results into some useful and transparent findings. To address such challenges, a thorough extension and generalization of well-known techniques such as regression analysis becomes essential and highly advantageous. In this paper, we extend the concept of regression models so that they can handle hybrid data coming from various sources which quite often exhibit diverse levels of data quality. The major objective of this study is to develop a sound vehicle of a hybrid data analysis, which helps in reducing the computing time, especially in cases of real-time data processing. We propose an efficient real-time fuzzy switching regression analysis based on a genetic algorithm-based fuzzy C-means associated with a convex hull-based fuzzy regression approach. The method enables us to deal with situations when one has to deal with heterogeneous data which were derived from various database sources (distributed databases). In the proposed design, we emphasize a pivotal role of the convex hull approach, which is essential to alleviate the limitations of linear programming when being used in modeling of real-time systems. O

show abstract

Greedy breadth-first-search based chain classification for nondestructive sizing of subsurface defects

Ahmed

Rao

Jayakumar

2016

Applied Soft Computing

View full text Add to dashboard Cite

Classification of Steam Generator Tube Defects for Real-Time Applications Using Eddy Current Test Data and Self-Organizing Maps

Cited by 6 publications

References 22 publications

Unsupervised Classification of Surface Defects in Wire Rod Production Obtained by Eddy Current Sensors

Unsupervised Classification of Surface Defects in Wire Rod Production Obtained by Eddy Current Sensors

A combination of genetic algorithm‐based fuzzy C‐means with a convex hull‐based regression for real‐time fuzzy switching regression analysis: application to industrial intelligent data analysis

Greedy breadth-first-search based chain classification for nondestructive sizing of subsurface defects

Contact Info

Product

Resources

About