EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Autoencoder-based compression and reconstruction of railway tunnel vibration data: a comparative study

Dianelys Vega Ruiz, Bernardo Lopes Poncetti, Marcos Massao Futai

Journal of Civil Structural Health Monitoring · 2027-02 · 17 (1)

DOI: 10.1007/s13349-026-01166-2 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Civil Structural Health Monitoring, online erschienen am 06.11.2026. Verfasst von Dianelys Vega Ruiz und 2 weiteren Autor:innen. Die Arbeit nennt 71 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Universidade De São Paulo.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

This paper provides a proof of concept comparative study of four autoencoder architectures for compressing acceleration signals collected from a real railway tunnel under realistic and noisy conditions. The evaluated models include a Single-layer Sparse Autoencoder (SAE), a Fully Connected Autoencoder (FC), a Convolutional Autoencoder (CAE), and a hybrid architecture that combines convolutional (CNN) and Long Short-Term Memory (LSTM) layers. All models are trained and tested on segmented time-series acceleration data acquired under real operating conditions, including measurement noise and operational variability typical of real-world tunnel structural health monitoring (SHM) signals. The performance of the models is evaluated is evaluated using multiple quantitative metrics, including Mean Absolute Error (MAE), Coefficient of Determination (R $$^{2}$$ ), Peak Signal-to-Noise Ratio (PSNR), and Cepstral Distance. A composite objective based on these metrics is employed to guide Bayesian hyperparameter optimization to select the optimal configurations. The best-performing model was a CNN-LSTM configuration. On average, CNN-LSTM models excelled at preserving temporal features (mean: MAE $$=$$ 0.38, R $$^{2} = $$ 0.73, PSNR $$=$$ 16.33 dB), while the SAE better preserved spectral characteristics (mean CD $$=$$ 0.72). These results demonstrate the feasibility of autoencoder-based compression for tunnel vibration data and highlight the importance of architecture selection according to SHM objectives.

Förderung

  • Universidade De São Paulo

Zitieren

APA

Ruiz, D. V., Poncetti, B. L., & Futai, M. M. (2027). Autoencoder-based compression and reconstruction of railway tunnel vibration data: a comparative study. Journal of Civil Structural Health Monitoring, 17(1). https://doi.org/10.1007/s13349-026-01166-2

Vancouver

Ruiz DV, Poncetti BL, Futai MM. Autoencoder-based compression and reconstruction of railway tunnel vibration data: a comparative study. Journal of Civil Structural Health Monitoring. 2027;17(1). doi:10.1007/s13349-026-01166-2

BibTeX
@article{Ruiz2027,
  author = {Ruiz, Dianelys Vega and Poncetti, Bernardo Lopes and Futai, Marcos Massao},
  title = {Autoencoder-based compression and reconstruction of railway tunnel vibration data: a comparative study},
  journal = {Journal of Civil Structural Health Monitoring},
  year = {2027},
  volume = {17},
  number = {1},
  doi = {10.1007/s13349-026-01166-2}
}
Kostenlos starten →