Open Access Peer-reviewed Research Article

Physics-Informed Digital Twin for Robust Inverse Parameter Identification under Sparse and Noisy Measurements

Article Sidebar

Schematic representation
Download PDF
Submitted   Jul 7, 2026
Published   Sep 11, 2026
Issue    Vol 5 No 2 (2026)
DOI   10.25082/RIMA.2026.02.002

Views

0

Downloads

0

Citations

PlumX Metrics

Main Article Content

Aswin Karkadakattil corresponding author
Independent Researcher, Kasaragod, Kerala 671314, India
Crossmark logo

Abstract

Reliable inverse identification of mechanical system parameters from sparse and noisy measurements remains a major challenge for digital twin applications because conventional identification methods are highly sensitive to measurement uncertainty and limited sensing data. This study presents a physics-informed digital twin framework that integrates the governing equation of motion with a neural network to simultaneously reconstruct system dynamics and estimate unknown mechanical parameters while maintaining physical consistency. The framework is developed for a single-degree-of-freedom mass–spring–damper system and jointly optimizes the displacement response together with the mass, damping, and stiffness using a two-stage Adam–L-BFGS optimization strategy. Its performance is evaluated through convergence analysis, basin stability, structural identifiability, ensemble-based uncertainty quantification, comparisons with least-squares, recursive least-squares, and data-driven neural networks, validation using independent literature-based parameter sets, and sensitivity analysis under signal-to-noise ratios ranging from 5 to 40 dB. The proposed framework reconstructs the system response with RMSE values of 0.013–0.021 while identifying mass, damping, and stiffness with relative errors of approximately 1.1%, 1.1%, and 1.3%, respectively. Compared with conventional and purely data-driven approaches, it achieves more accurate parameter estimation, lower reconstruction error (RMSE = 0.016), consistent convergence across multiple random initializations, and reliable 95% confidence intervals for prediction uncertainty. Unlike previous studies that primarily demonstrate inverse parameter recovery using physics-informed neural networks, this work provides a comprehensive reliability-oriented evaluation by integrating optimization robustness, structural identifiability, uncertainty quantification, comparative benchmarking, and multi-noise validation within a single physics-informed digital twin framework, improving the reliability and interpretability of inverse parameter identification under limited sensing conditions.

Keywords
physics-informed neural networks, digital twin, inverse problems, system identification, uncertainty quantification, Mechanical dynamics

Article Details

How to Cite
Karkadakattil, A. (2026). Physics-Informed Digital Twin for Robust Inverse Parameter Identification under Sparse and Noisy Measurements. Research on Intelligent Manufacturing and Assembly, 5(2), 394-414. https://doi.org/10.25082/RIMA.2026.02.002
Creative Commons License

This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.

References

  1. Ceccarelli D. Bayesian physics-informed neural networks for inverse uncertainty quantification problems in cardiac electrophysiology. PhD Thesis/Technical Report, 2019.
  2. Holland JR, Baeder JD, Duraisamy K. Field Inversion and Machine Learning With Embedded Neural Networks: Physics-Consistent Neural Network Training. AIAA Aviation 2019 Forum. Published online June 14, 2019. https://doi.org/10.2514/6.2019-3200
  3. Castellani A, Schmitt S, Squartini S. Real-World Anomaly Detection by Using Digital Twin Systems and Weakly Supervised Learning. IEEE Transactions on Industrial Informatics. 2021, 17(7): 4733-4742. https://doi.org/10.1109/tii.2020.3019788
  4. Chen Y, Lu L, Karniadakis GE, et al. Physics-informed neural networks for inverse problems in nano-optics and metamaterials. Optics Express. 2020, 28(8): 11618. https://doi.org/10.1364/oe.384875
  5. Lu L, Pestourie R, Yao W, et al. Physics-Informed Neural Networks with Hard Constraints for Inverse Design. SIAM Journal on Scientific Computing. 2021, 43(6): B1105-B1132. https://doi.org/10.1137/21m1397908
  6. Cai S, Mao Z, Wang Z, et al. Physics-informed neural networks (PINNs) for fluid mechanics: a review. Acta Mechanica Sinica. 2021, 37(12): 1727-1738. https://doi.org/10.1007/s10409-021-01148-1
  7. Gao C, Park H, Easwaran A. An anomaly detection framework for digital twin driven cyber-physical systems. Proceedings of the ACM/IEEE 12th International Conference on Cyber-Physical Systems. Published online May 19, 2021: 44-54. https://doi.org/10.1145/3450267.3450533
  8. Chen J, Dai Z, Yang Z, et al. An Improved Tandem Neural Network Architecture for Inverse Modeling of Multicomponent Reactive Transport in Porous Media. Water Resources Research. 2021, 57(12). https://doi.org/10.1029/2021wr030595
  9. Wang N, Chang H, Zhang D. Deep‐Learning‐Based Inverse Modeling Approaches: A Subsurface Flow Example. Journal of Geophysical Research: Solid Earth. 2021, 126(2). https://doi.org/10.1029/2020jb020549
  10. Depina I, Jain S, Mar Valsson S, et al. Application of physics-informed neural networks to inverse problems in unsaturated groundwater flow. Georisk: Assessment and Management of Risk for Engineered Systems and Geohazards. 2021, 16(1): 21-36. https://doi.org/10.1080/17499518.2021.1971251
  11. Jiang J, Li H, Mao Z, et al. A digital twin auxiliary approach based on adaptive sparse attention network for diesel engine fault diagnosis. Scientific Reports. 2022, 12(1). https://doi.org/10.1038/s41598-021-04545-5
  12. Wang J, Moreira J, Cao Y, et al. Time-Variant Digital Twin Modeling through the Kalman-Generalized Sparse Identification of Nonlinear Dynamics. 2022 American Control Conference (ACC). Published online June 8, 2022: 5217-5222. https://doi.org/10.23919/acc53348.2022.9867786
  13. Bharadwaja BVSS, Nabian MA, Sharma B, et al. Physics-Informed Machine Learning and Uncertainty Quantification for Mechanics of Heterogeneous Materials. Integrating Materials and Manufacturing Innovation. 2022, 11(4): 607-627. https://doi.org/10.1007/s40192-022-00283-2
  14. Genzel M, Macdonald J, Marz M. Solving Inverse Problems With Deep Neural Networks – Robustness Included? IEEE Transactions on Pattern Analysis and Machine Intelligence. 2023, 45(1): 1119-1134. https://doi.org/10.1109/tpami.2022.3148324
  15. Kapoor T, Wang H, Núnez A, et al. Physics-Informed Neural Networks for Solving Forward and Inverse Problems in Complex Beam Systems. IEEE Transactions on Neural Networks and Learning Systems. 2024, 35(5): 5981-5995. https://doi.org/10.1109/tnnls.2023.3310585
  16. Baldan M, Di Barba P, Lowther DA. Physics-Informed Neural Networks for Inverse Electromagnetic Problems. IEEE Transactions on Magnetics. 2023, 59(5): 1-5. https://doi.org/10.1109/tmag.2023.3247023
  17. Wang J, Moreira J, Cao Y, et al. Simultaneous digital twin identification and signal-noise decomposition through modified generalized sparse identification of nonlinear dynamics. Computers & Chemical Engineering. 2023, 177: 108294. https://doi.org/10.1016/j.compchemeng.2023.108294
  18. Mao C, Jin Y. Uncertainty quantification study of the physics-informed machine learning models for critical heat flux prediction. Progress in Nuclear Energy. 2024, 170: 105097. https://doi.org/10.1016/j.pnucene.2024.105097
  19. Michek NE, Mehta P, Huebsch WW. Flight Dynamic Uncertainty Quantification Modeling Using Physics-Informed Neural Networks. AIAA Journal. 2024, 62(11): 4234-4246. https://doi.org/10.2514/1.j063992
  20. Liu L, Liu S, Xie H, et al. Discontinuity Computing Using Physics-Informed Neural Networks. Journal of Scientific Computing. 2023, 98(1). https://doi.org/10.1007/s10915-023-02412-1
  21. Xu X, Paneru S, Russcher SA, et al. Physics-Guided Bayesian Neural Networks and Their Application in ODE Problems. ASME 2024 Verification, Validation, and Uncertainty Quantification Symposium. Published online May 15, 2024. https://doi.org/10.1115/vvuq2024-122961
  22. Oddiraju M, Penumatsa BV, Amin D, et al. Exploring Efficient Quantification of Modeling Uncertainties With Differentiable Physics-Informed Machine Learning Architectures. Volume 3B: 51st Design Automation Conference (DAC). Published online August 17, 2025. https://doi.org/10.1115/detc2025-169099
  23. Shi Y, Wei P, Feng K, et al. A survey on machine learning approaches for uncertainty quantification of engineering systems. Machine Learning for Computational Science and Engineering. 2025, 1(1). https://doi.org/10.1007/s44379-024-00011-x
  24. Teloli R, Bigot M, Coelho L, et al. Physics-informed neural networks for inverse problems in structural dynamics. Shull PJ, Yu T, Gyekenyesi AL, Wu HF, eds. Nondestructive Characterization and Monitoring of Advanced Materials, Aerospace, Civil Infrastructure, and Transportation XVIII. Published online May 9, 2024: 19. https://doi.org/10.1117/12.3010918
  25. Wang J, Moreira J, Cao Y, et al. Neural network and Sparse identification of Nonlinear Dynamics Integrated Algorithm for Digital Twin identification. IFAC-PapersOnLine. 2023, 56(2): 6921-6926. https://doi.org/10.1016/j.ifacol.2023.10.503
  26. Sahin T, von Danwitz M, Popp A. Solving forward and inverse problems of contact mechanics using physics-informed neural networks. Advanced Modeling and Simulation in Engineering Sciences. 2024, 11(1). https://doi.org/10.1186/s40323-024-00265-3
  27. Karkadakattil A. A physics-informed intelligent digital twin using multi-task CNN–LSTM for acoustic emission-based fault prognostics in safety-critical systems. Journal of Intelligent Manufacturing and Special Equipment. 2026, 7(1): 33-58. https://doi.org/10.1108/jimse-10-2025-0023