Railway Wear: Prediction Using FCNN and Diagnosis Using PCA+SVM
Vehicles · 03.11.2026 · 8 (11) · 286
DOI: 10.3390/vehicles8110286 ↗Kurz gefasst
Fachartikel in der Zeitschrift Vehicles, online erschienen am 03.11.2026. Verfasst von Tania Elizabeth Sandoval Valencia und 6 weiteren Autor:innen. Die Arbeit nennt 38 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: SECIHTI, Postdoctoral stays in Mexico.
Abstract (Originalsprache)
The main challenge in railway predictive maintenance is balancing quantitative precision with interpretability, given the scarcity of experimental data linking dynamics and wear. In this study, a 1:20 scale test rig equipped with accelerometers, a gyroscope, and microscopy generated a multimodal dataset comprising 2460 cycles, synchronizing vibration data (1 kHz) with sequential images of the rolling surface. Two models were developed and compared: (1) Principal Component Analysis and Support Vector Machines (PCA+SVM) to classify four wear states, and (2) a Fully Connected Neural Network (FCNN) for Remaining Useful Life (RUL) regression. The PCA+SVM model achieved 84% accuracy and, through factor loadings, identified high-frequency energy (200–500 Hz) and kurtosis as physical precursors to fatigue. The FCNN model yielded a Mean Absolute Error (MAE) of 100 cycles (NMAE = 4.06%). Together, these results show that PCA+SVM is well-suited for interpretable diagnostics, while FCNN excels at quantitative forecasting, laying the groundwork for a hybrid predictive system.
Förderung
- SECIHTI
- Postdoctoral stays in Mexico
Zitieren
APA
Valencia, T. E. S., Hurtado-Hurtado, G., Ramírez, J. E. L., Basaldúa-Sánchez, J. E., Flores, G. M. S., Viadero-Monasterio, F., & Jáuregui-Correa, J. C. (2026). Railway Wear: Prediction Using FCNN and Diagnosis Using PCA+SVM. Vehicles, 8(11), 286. https://doi.org/10.3390/vehicles8110286
Vancouver
Valencia TES, Hurtado-Hurtado G, Ramírez JEL, Basaldúa-Sánchez JE, Flores GMS, Viadero-Monasterio F, et al. Railway Wear: Prediction Using FCNN and Diagnosis Using PCA+SVM. Vehicles. 2026;8(11):286. doi:10.3390/vehicles8110286
BibTeX
@article{Valencia2026,
author = {Valencia, Tania Elizabeth Sandoval and Hurtado-Hurtado, Gerardo and Ramírez, Jose Emiliano López and Basaldúa-Sánchez, J. E. and Flores, Gabriela Montserat Sánchez and Viadero-Monasterio, Fernando and Jáuregui-Correa, Juan Carlos},
title = {Railway Wear: Prediction Using FCNN and Diagnosis Using PCA+SVM},
journal = {Vehicles},
year = {2026},
volume = {8},
number = {11},
pages = {286},
doi = {10.3390/vehicles8110286}
}
EUVIMED