Intelligent Classification Method for Rail Defects in Magnetic Flux Leakage Testing Based on Feature Selection and Parameter Optimization
This study addresses the critical challenge of insufficient classification accuracy for different defect signals in rail magnetic flux leakage (MFL) detection by proposing an enhanced intelligent classification framework based on particle swarm optimized radial basis function neural network (PSO-RBF...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-06-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/25/13/3962 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849703764998488064 |
|---|---|
| author | Kailun Ji Ping Wang Yinliang Jia |
| author_facet | Kailun Ji Ping Wang Yinliang Jia |
| author_sort | Kailun Ji |
| collection | DOAJ |
| description | This study addresses the critical challenge of insufficient classification accuracy for different defect signals in rail magnetic flux leakage (MFL) detection by proposing an enhanced intelligent classification framework based on particle swarm optimized radial basis function neural network (PSO-RBF). Three key innovations drive this research: (1) A dynamic PSO algorithm incorporating adaptive learning factors and nonlinear inertia weight for precise RBF parameter optimization; (2) A hierarchical feature processing strategy combining mutual information selection with correlation-based dimensionality reduction; (3) Adaptive model architecture adjustment for small-sample scenarios. Experimental validation shows breakthrough performance: 87.5% accuracy on artificial defects (17.5% absolute improvement over conventional RBF), with macro-F1 = 0.817 and MCC = 0.733. For real-world limited samples (100 sets), adaptive optimization achieved 80% accuracy while boosting minority class (“spalling”) F1-score by 0.25 with 50% false alarm reduction. The optimized PSO-RBF demonstrates superior capability in extracting MFL signal patterns, particularly for discriminating abrasions, spalling, indentations, and shelling defects, setting a new benchmark for industrial rail inspection. |
| format | Article |
| id | doaj-art-6aab88d8bbef407884f4bc97e9033da5 |
| institution | DOAJ |
| issn | 1424-8220 |
| language | English |
| publishDate | 2025-06-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj-art-6aab88d8bbef407884f4bc97e9033da52025-08-20T03:17:08ZengMDPI AGSensors1424-82202025-06-012513396210.3390/s25133962Intelligent Classification Method for Rail Defects in Magnetic Flux Leakage Testing Based on Feature Selection and Parameter OptimizationKailun Ji0Ping Wang1Yinliang Jia2College of Automation Engineering, Nanjing University of Aeronautics and Astronautics, Nanjing 210016, ChinaCollege of Automation Engineering, Nanjing University of Aeronautics and Astronautics, Nanjing 210016, ChinaCollege of Automation Engineering, Nanjing University of Aeronautics and Astronautics, Nanjing 210016, ChinaThis study addresses the critical challenge of insufficient classification accuracy for different defect signals in rail magnetic flux leakage (MFL) detection by proposing an enhanced intelligent classification framework based on particle swarm optimized radial basis function neural network (PSO-RBF). Three key innovations drive this research: (1) A dynamic PSO algorithm incorporating adaptive learning factors and nonlinear inertia weight for precise RBF parameter optimization; (2) A hierarchical feature processing strategy combining mutual information selection with correlation-based dimensionality reduction; (3) Adaptive model architecture adjustment for small-sample scenarios. Experimental validation shows breakthrough performance: 87.5% accuracy on artificial defects (17.5% absolute improvement over conventional RBF), with macro-F1 = 0.817 and MCC = 0.733. For real-world limited samples (100 sets), adaptive optimization achieved 80% accuracy while boosting minority class (“spalling”) F1-score by 0.25 with 50% false alarm reduction. The optimized PSO-RBF demonstrates superior capability in extracting MFL signal patterns, particularly for discriminating abrasions, spalling, indentations, and shelling defects, setting a new benchmark for industrial rail inspection.https://www.mdpi.com/1424-8220/25/13/3962rail defect detectionmagnetic flux leakage (MFL) testingfeature selectionintelligent classificationmodel optimizationimbalanced data |
| spellingShingle | Kailun Ji Ping Wang Yinliang Jia Intelligent Classification Method for Rail Defects in Magnetic Flux Leakage Testing Based on Feature Selection and Parameter Optimization Sensors rail defect detection magnetic flux leakage (MFL) testing feature selection intelligent classification model optimization imbalanced data |
| title | Intelligent Classification Method for Rail Defects in Magnetic Flux Leakage Testing Based on Feature Selection and Parameter Optimization |
| title_full | Intelligent Classification Method for Rail Defects in Magnetic Flux Leakage Testing Based on Feature Selection and Parameter Optimization |
| title_fullStr | Intelligent Classification Method for Rail Defects in Magnetic Flux Leakage Testing Based on Feature Selection and Parameter Optimization |
| title_full_unstemmed | Intelligent Classification Method for Rail Defects in Magnetic Flux Leakage Testing Based on Feature Selection and Parameter Optimization |
| title_short | Intelligent Classification Method for Rail Defects in Magnetic Flux Leakage Testing Based on Feature Selection and Parameter Optimization |
| title_sort | intelligent classification method for rail defects in magnetic flux leakage testing based on feature selection and parameter optimization |
| topic | rail defect detection magnetic flux leakage (MFL) testing feature selection intelligent classification model optimization imbalanced data |
| url | https://www.mdpi.com/1424-8220/25/13/3962 |
| work_keys_str_mv | AT kailunji intelligentclassificationmethodforraildefectsinmagneticfluxleakagetestingbasedonfeatureselectionandparameteroptimization AT pingwang intelligentclassificationmethodforraildefectsinmagneticfluxleakagetestingbasedonfeatureselectionandparameteroptimization AT yinliangjia intelligentclassificationmethodforraildefectsinmagneticfluxleakagetestingbasedonfeatureselectionandparameteroptimization |