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

Aug 3, 2026

Upload by Eliana

Feature selection strongly influences learning accuracy, computation cost, and interpretability in machine learning systems. This study develops a prediction and analysis framework for steam generator tube crack dimensions using 22 gray-level co-occurrence matrix features extracted from magnetic flux leakage images. Model performance is assessed with R2 and RMSE on training and testing sets, while F Score and mutual information ranking prioritize features. A Taguchi design and ANOVA evaluate effects of learning models and selection strategies. A dynamic population gray wolf algorithm (DPGWO) selects optimal feature subsets and combinations, and Pareto optimal solutions are determined via Deng’s method, validated against other optimizers using Friedman tests and multiple performance indicators.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages33

Reading Time1 h 23 min

AccessFree to download

Introduction

Background and Need for Inspection

Non-destructive Testing and Magnetic Flux Leakage

Machine Learning for Crack Characterization

Methodology (from Abstract)

Feature Extraction and Prioritization

Model Evaluation Metrics

Taguchi Design and ANOVA

DPGWO-based Feature Selection and Pareto Optimization

Comparative Validation

applied sciences Article DPGWO Based Feature Selection Machine Learning Model for Prediction of Crack Dimensions in Steam Generator Tubes Mathias Vijay Albert William 1, Subramanian Ramesh 2, Robert Cep 3, *, Siva Kumar Mahalingam 4 and Muniyandy Elangovan 5,6, * Citation: William, M.V.A.; Ramesh, S.; Cep, R.; Mahalingam, S.K.; Elangovan, M. DPGWO Based Feature Selection Machine Learning Model for Prediction of Crack Dimensions in Steam Generator Tubes. Appl. Sci. 2023, 13, 8206 . [https://doi.org/10.3390/](https://doi.org/10.3390/)[ ](https://doi.org/10.3390/)app13148206 Academic Editor: Ki-Yong Oh Received: 21 June 2023 Revised: 11 July 2023 Accepted: 13 July 2023 Published: 14 July 2023 Copyright: © 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license ([https://](https://)[ ](https://)[creativecommons.org/licenses/by/](creativecommons.org/licenses/by/)[ ](creativecommons.org/licenses/by/)[4.0/](4.0/)) . 1 Department of Electronics and Communication Engineering, Vel Tech Rangarajan Dr. Sagunthala R&D Institute of Science and Technology, Avadi 600062, India 2 Department of Electrical and Electronics Engineering, Vel Tech Rangarajan Dr. Sagunthala R&D Institute of Science and Technology, Avadi 600062, India 3 Department of Machining, Assembly and Engineering Metrology, Faculty of Mechanical Engineering, VSB-Technical University of Ostrava, 70800 Ostrava, Czech Republic 4 Department of Mechanical Engineering, Vel Tech Rangarajan Dr. Sagunthala R&D Institute of Science and Technology, Avadi 600062, India 5 Department of Biosciences, Saveetha School of Engineering, Saveetha Nagar, Thandalam 602105, India 6 Department of R&D, Bond Marine Consultancy, London EC1V 2NX, UK * Correspondence: [robert.cep@vsb.cz](robert.cep@vsb.cz) (R.C.); [muniyandy.e@gmail.com](muniyandy.e@gmail.com) (M.E.) Abstract: 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. Keywords: machine learning model; feature selection methods; optimization algorithms; Friedman test; Deng's methods; performance indicators 1. Introduction In nuclear power plants, critical components such as steam generator tubes (SGT), feed water heaters, and pressure vessels have stringent design requirements due to the high temperature, pressure, and radiation-exposing environment, which induce stress corrosion cracking, pitting, fouling, and mechanical fretting [1] . Apart from that, steam generator tube rupture (SGTR) leads to the inevitable rele

FAQs

What data representation is used to model crack dimensions in steam generator tubes?

The approach uses 22 gray-level co-occurrence matrix features extracted from magnetic flux leakage images of tube cracks.

How are feature subsets prioritized before training the prediction model?

Features are prioritized using F Score and Mutual Information Score methods to guide selection toward more informative inputs.

How does DPGWO support the final selection of feature combinations?

DPGWO selects the best feature sets and their combinations, and Pareto optimal solutions are chosen using Deng’s method to address contradictory performance metrics.

Aug 3, 2026

Upload by Eliana