EUVIMEDEUVIMED entdecken
Crossref

A Digital Twin Framework With Adaptive Learning for Resilient Structural Health Monitoring

Vijay Kumar, Munish Bhatia

Concurrency and Computation: Practice and Experience · 2026-11 · 38 (22)

DOI: 10.1002/cpe.71025 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Concurrency and Computation: Practice and Experience, online erschienen am 08.11.2026. Verfasst von Vijay Kumar und 1 weiteren Autor:innen. Die Arbeit nennt 28 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Digital Twin (DT) technology facilitates real‐time virtual simulation of physical infrastructure and has proven to be a viable answer to structural health monitoring (SHM), especially during times of disaster. The current study introduces a DT‐based intelligent SHM system, which combines convolutional neural network (CNN)‐based classification and gated recurrent unit (GRU)‐based temporal prediction to evaluate the structural health accurately. The framework takes multi‐sensor data and categorises it as healthy or unhealthy using CNN and predicts health dynamics using GRU in a DT environment. It uses moisture, vibration, strain, and crack growth information to simulate the realistic structural behavior. The framework is tested using four datasets (D1–D4) consisting of 35,041, 40,412, 50,312, and 60,215 instances, respectively. Findings indicate that the CNN classifier has an average‐accuracy of 93.54%, sensitivity of 90.88%, specificity of 95.26% and F1‐score of 92.12% in each dataset. The GRU model has a good predictive performance with a average accuracy of more than 97.78% and low prediction errors (mean absolute error (MAE) = 0.0049, root mean squared error (RMSE) = 0.11) and high Pearson correlation coefficients (r) (0.994). The proposed framework introduces a temporal abstraction‐based Digital Twin SHM model with integrated classification, prediction, and decision‐level intelligence for real‐time structural health assessment. The proposed DT‐enabled framework demonstrates the potential to facilitate structural health evaluation, early warning generation, and predictive maintenance, thereby improving infrastructure resilience in disaster‐prone environments.

Zitieren

APA

Kumar, V., & Bhatia, M. (2026). A Digital Twin Framework With Adaptive Learning for Resilient Structural Health Monitoring. Concurrency and Computation: Practice and Experience, 38(22). https://doi.org/10.1002/cpe.71025

Vancouver

Kumar V, Bhatia M. A Digital Twin Framework With Adaptive Learning for Resilient Structural Health Monitoring. Concurrency and Computation: Practice and Experience. 2026;38(22). doi:10.1002/cpe.71025

BibTeX
@article{Kumar2026,
  author = {Kumar, Vijay and Bhatia, Munish},
  title = {A Digital Twin Framework With Adaptive Learning for Resilient Structural Health Monitoring},
  journal = {Concurrency and Computation: Practice and Experience},
  year = {2026},
  volume = {38},
  number = {22},
  doi = {10.1002/cpe.71025}
}
Kostenlos starten →