Magnetic Flux Leakage (MFL) based Defect Characterization of Steam Generator Tubes using Artificial Neural Networks

Daniel, Jackson; Abudhahir, A.; Paulin, J. Janet

doi:10.4283/jmag.2017.22.1.034

Cited by 11 publications

(13 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Overfitting of models is one of the major problems that will reduce the application of ML in various fields. The problem considered in this work will be developing an accurate ML model with a smaller number of features to predict the crack dimensions from 22 GLCM features extracted from 105 MFL crack images presented by Daniel et al [12].…”

Section: Problem Statementmentioning

confidence: 99%

See 1 more Smart Citation

DPGWO Based Feature Selection Machine Learning Model for Prediction of Crack Dimensions in Steam Generator Tubes

et al. 2023

View full text Add to dashboard Cite

The selection of an appropriate number of features and their combinations will play a major role in improving the learning accuracy, computation cost, and understanding of machine learning models. In this present work, 22 gray-level co-occurrence matrix features extracted from magnetic flux leakage images captured in steam generator tubes’ cracks are considered for developing a machine learning model to predict and analyze crack dimensions in terms of their length, depth, and width. The performance of the models is examined by considering R2 and RMSE values calculated using both training and testing data sets. The F Score and Mutual Information Score methods have been applied to prioritize the features. To analyze the effect of different machine learning models, their number of features, and their selection methods, a Taguchi experimental design has been implemented and an analysis of variance test has been conducted. The dynamic population gray wolf algorithm (DPGWO) has been adopted to select the best features and their combinations. Due to the two contradictory natures of performance metrics, Pareto optimal solutions are considered, and the best one is obtained using Deng’s method. The effectiveness of DPGWO is proved by comparing its performance with Grey Wolf Optimization and Moth Flame Optimization algorithms using the Friedman test and performance indicators, namely inverted generational distance and spacing.

show abstract

Section: Problem Statementmentioning

confidence: 99%

“…The experimental setup for the measurement of MFL is detailed in Suresh et al [10]. Daniel et al [12] designed an ANN model to forecast the SGT's defect in terms of the length, breadth, and depth of the crack by providing the gray-level co-occurrence matrix (GLCM) information extracted from the MFL images.…”

Section: Introductionmentioning

confidence: 99%

DPGWO Based Feature Selection Machine Learning Model for Prediction of Crack Dimensions in Steam Generator Tubes

et al. 2023

View full text Add to dashboard Cite

show abstract

“…Magnetic flux leakage (MFL) is the most popular magnetic sensing method used in this study to investigate defect features, including width, depth, and area. In the past, magnetic flux leakage (MFL) has been used for testing pipelines [ 21 , 22 , 23 , 24 , 25 ] and inspecting plates [ 26 , 27 ], tubes [ 28 , 29 , 30 ], wire ropes [ 31 , 32 , 33 , 34 , 35 ], rail tracks [ 36 ], tank floors [ 37 , 38 ], and suspension bridge stay cables [ 39 ] made up of ferromagnetic material.…”

Section: Introductionmentioning

confidence: 99%

Evaluation of the Size of a Defect in Reinforcing Steel Using Magnetic Flux Leakage (MFL) Measurements

Yousaf,

Harseno,

Kee

et al. 2023

Sensors

View full text Add to dashboard Cite

This study aimed to evaluate 2D magnetic flux leakage (MFL) signals (Bx, By) in D19-size reinforcing steel with several defect conditions. The magnetic flux leakage data were collected from the defected and new specimens using an economically designed test setup incorporating permanent magnets. A two-dimensional finite element model was numerically simulated using COMSOL Multiphysics to validate the experimental tests. Based on the MFL signals (Bx, By), this study also intended to improve the ability to analyze defect features such as width, depth, and area. Both the numerical and experimental results indicated a high cross-correlation with a median coefficient of 0.920 and a mean coefficient of 0.860. Using signal information to evaluate defect width, the x-component (Bx) bandwidth was found to increase with increasing defect width and the y-component (By) amplitude rise with increasing depth. In this two-dimensional MFL signal study, both parameters of the two-dimensional defects (width and depth) affected each other and could not be evaluated individually. The defect area was estimated from the overall variation in the signal amplitude of the magnetic flux leakage signals with the x-component (Bx). The defect areas showed a higher regression coefficient (R2 = 0.9079) for the x-component (Bx) amplitude from the 3-axis sensor signal. It was determined that defect features are positively correlated with sensor signals.

show abstract

“…ey used COMSOL Multiphysics 4.3a software to model the rectangular notch defect on the outer surface of the steam generator tube. His research process is too complicated [3]. Carvalho et al believed that topological invariants can represent the Hamiltonian, thereby predicting the existence of gap modes for topological protection.…”

Section: Introductionmentioning

confidence: 99%

Diagnosis Model of Volleyball Skills and Tactics Based on Artificial Neural Network

Jiang

Zhao

Jin

2021

Mobile Information Systems

View full text Add to dashboard Cite

With the development of China’s sports industry, the technical and tactical level of the team is required to be higher and higher. This study mainly discusses the diagnostic model of volleyball technique and tactics based on artificial neural network. With the help of the correlation function in Matlab neural network toolbox, in the training process of volleyball technical and tactical evaluation neural network, the sample data of volleyball technical evaluation index is repeatedly simulated and studied, and finally the network parameters with the minimum error and the highest accuracy are saved as the network model for subsequent verification and evaluation. The middle layer is the hidden layer, which makes the network approach the result of volleyball experts’ evaluation of the technology by adjusting the weights of neurons. The last layer is the output layer, which outputs the actual evaluation results of volleyball experts on the technology. Through repeated training and comparison of input samples, the maximum number of training times of BP network for volleyball technical and tactical evaluation is determined to be 32. Some common experience of estimating hidden node number is provided by trial-and-error method. On this basis, the number of hidden nodes to minimize the network error is finally determined to be 4 through repeated training and comparison. In the process of network diagnosis, the average difference between the evaluation score of network output and the score of actual experts is less than 1%, which reaches a very high precision. It shows that the volleyball skill evaluation model based on BP neural network is feasible in technology and the result is relatively reliable.

show abstract

Magnetic Flux Leakage (MFL) based Defect Characterization of Steam Generator Tubes using Artificial Neural Networks

Cited by 11 publications

References 12 publications

DPGWO Based Feature Selection Machine Learning Model for Prediction of Crack Dimensions in Steam Generator Tubes

DPGWO Based Feature Selection Machine Learning Model for Prediction of Crack Dimensions in Steam Generator Tubes

Evaluation of the Size of a Defect in Reinforcing Steel Using Magnetic Flux Leakage (MFL) Measurements

Diagnosis Model of Volleyball Skills and Tactics Based on Artificial Neural Network

Contact Info

Product

Resources

About