EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Feature-Level Decoupling of Reinforced-Concrete Parameters from Sparse MFL Data via Three-Layer Sensing

Paweł Karol Frankowski, Cindy Almosura Lasco iD, Alan Karl Syting Gatdula, Marcin Mąka iD, Ryszard Daniel Ziętek iD, Radosław Gordon iD, Paweł Owczarek iD, Patryk Pocheć iD, Tomasz Stawicki iD, Sebastian Matysik iD

Sensors · 07.11.2026 · 26 (21) · 7049

DOI: 10.3390/s26217049 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, online erschienen am 07.11.2026. Verfasst von Paweł Karol Frankowski und 9 weiteren Autor:innen. Beteiligte Einrichtung: Department of Information and Communication Technologies, Faculty of Computer Science and Telecommunications, Maritime University of Szczecin, Wały C…. Die Arbeit nennt 84 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

This methodological paper presents a data-efficient, explainable artificial intelligence (XAI) pipeline for non-destructive testing (NDT) of reinforced concrete (RC) under sparse-data conditions, addressing three closely connected challenges: cross-parameter interference in magnetic flux leakage (MFL) signals, scarce training data, and black-box opacity. Data scarcity is a recognized constraint across AI-driven NDT and is structural in reinforced concrete (costly labeling, absent benchmarks, small-data learning missing from the reviewed electromagnetic literature). This scarcity has two consequences: black-box models can generalize poorly on sparse data, and where errors are likely, engineers hesitate to trust or assume liability for opaque predictions. We present a glass-box framework identifying rebar diameter (D), cover thickness (h), and steel class through feature-level decoupling grounded in measurement physics. Transparency and accuracy are treated as equally vital NDT dimensions. Amplitude–Correlation–Offset (ACO) decomposition of a three-layer magnetoresistive transducer isolates parameter-specific attributes: Bx offset classifies steel class (95% confidence), genetic-algorithm fitting localizes peaks to 0.1 mm for D (10–12 mm), and inter-layer amplitude ratios estimate h (20–60 mm). Rough-set analysis quantifies rule confidence and uncertainty zones. Proof-of-concept validation (20 specimens and 400 measurements) shows that a few reference configurations suffice for interpretable identification, enabling explainable artificial intelligence (XAI) in sparse-data NDT and lowering the technology-adoption barrier in safety-critical inspection.

Schlagwörter

Stichwörter der Autor:innen

association rule analysisexplainable artificial intelligencefew-shot data learningmagnetic flux leakagenon-destructive testingparameter decouplingreinforced concretesignal processingsparse datatechnology acceptance

Autor:innen und Institutionen

  • Department of Information and Communication Technologies, Faculty of Computer Science and Telecommunications, Maritime University of Szczecin, Wały Chrobrego 1-2, 70-500 Szczecin, Poland.
  • Faculty of Computing, Engineering and Technology, Davao Oriental State University, Guang-guang, Dahican, City of Mati 8200, Davao Oriental, Mindanao, Philippines.
  • Department of Robotics and Control, Faculty of Mechatronics and Electrical Engineering, Maritime University of Szczecin, Willowa 2, 71-650 Szczecin, Poland.
  • Department of Electrical Engineering and Power Electronics, Faculty of Mechatronics and Electrical Engineering, Maritime University of Szczecin, Willowa 2, 71-650 Szczecin, Poland.
  • Student Science Club KN BIT, Faculty of Computer Science and Telecommunications, Maritime University of Szczecin, Wały Chrobrego 1-2, 70-500 Szczecin, Poland.
  • Department of Engineering of Renewable Energy Sources, Faculty of Environmental Management and Agriculture, West Pomeranian University of Technology in Szczecin, Papieza Pawla VI Street 1, 71-459 Szczecin, Poland.

ORCID-Profile: Cindy Almosura Lasco, Marcin Mąka, Ryszard Daniel Ziętek, Radosław Gordon, Paweł Owczarek, Patryk Pocheć, Tomasz Stawicki, Sebastian Matysik

Zitieren

APA

Frankowski, P. K., Lasco, C. A., Gatdula, A. K. S., Mąka, M., Ziętek, R. D., Gordon, R., Owczarek, P., Pocheć, P., Stawicki, T., & Matysik, S. (2026). Feature-Level Decoupling of Reinforced-Concrete Parameters from Sparse MFL Data via Three-Layer Sensing. Sensors, 26(21), 7049. https://doi.org/10.3390/s26217049

Vancouver

Frankowski PK, Lasco CA, Gatdula AKS, Mąka M, Ziętek RD, Gordon R, et al. Feature-Level Decoupling of Reinforced-Concrete Parameters from Sparse MFL Data via Three-Layer Sensing. Sensors. 2026;26(21):7049. doi:10.3390/s26217049

BibTeX
@article{Frankowski2026,
  author = {Frankowski, Paweł Karol and Lasco, Cindy Almosura and Gatdula, Alan Karl Syting and Mąka, Marcin and Ziętek, Ryszard Daniel and Gordon, Radosław and Owczarek, Paweł and Pocheć, Patryk and Stawicki, Tomasz and Matysik, Sebastian},
  title = {Feature-Level Decoupling of Reinforced-Concrete Parameters from Sparse MFL Data via Three-Layer Sensing},
  journal = {Sensors},
  year = {2026},
  volume = {26},
  number = {21},
  pages = {7049},
  doi = {10.3390/s26217049}
}
Kostenlos starten →