“Frequency-Domain Physics-Informed Neural Networks for Parameter Estimation”
Authors: Jawad Tariq, Tuomo Lindh and Niko Nevaranta,Affiliation: Lappeenranta University of Technology
Reference: 2026, Vol 47, No 2, pp. 39-50.
Keywords: frequency domain, parameter estimation, physics-informed neural network (PINN), system identification
Abstract: While physics-informed neural networks (PINNs) are commonly employed for time-domain problems, this work extends their application to the frequency domain for parameter estimation. The framework integrates the governing equation into the neural network (NN) training loss function by using a composite complex domain formulation to enforce data consistency, minimize physics residuals, and respect parameter bounds. The methodology is presented as a transfer function-based formulation, whose parameters can be estimated from available data. The effectiveness of the methodology is validated through both simulation and experimental datasets on axial active magnetic bearing (AMB) systems, which are open-loop unstable and thus make frequency-domain parameter estimation particularly relevant. The results show excellent agreement between measured and predicted responses. Furthermore, the proposed approach is compared with an existing frequency-domain identification approach. PINN-based estimation of physical parameters is closer to reference values due to physics residual and parameter bounds. The findings validate the proposed methodology, which indicates that the PINN-based frequency-domain framework can accurately estimate system parameters, demonstrating its accuracy and robustness for physics-constrained learning in the frequency domain.
PDF (1177 Kb)
DOI: 10.4173/mic.2026.2.1References:
[1] Abubakar, I., Putkonen, A., Rehtla, M., Lindh, T., Derammelaere, S., and Nevaranta, N. (2024). Optimizing PID Controller Design for Rotor Systems Suspended by Active Magnetic Bearings, Modeling, Identification and Control. 45(3):105--114. doi:10.4173/mic.2024.3.3
[2] Aguirre, L.A., Barbosa, B.H., and Braga, A.P. (2010). Prediction and simulation errors in parameter estimation for nonlinear systems, Mechanical Systems and Signal Processing. 24(8):2855--2867. doi:10.1016/j.ymssp.2010.05.003
[3] Dey, S. and Mallik, A. (2025). Physics informed neural network—estimated circuit parameter adaptive modulation of dab, IEEE Transactions on Power Electronics. 40(10):14821--14841. doi:10.1109/TPEL.2025.3574873
[4] Jastrzebski, R.P. and Liukkonen, O. (2023). Analysis of a segmented axial active magnetic bearing for multi-mw compressor applications, IEEE/ASME Transactions on Mechatronics. 28(5):2799--2809. doi:10.1109/TMECH.2023.3254812
[5] Kapoor, T., Wang, H., Nunez, A., and Dollevoet, R. (2024). 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. doi:10.1109/TNNLS.2023.3310585
[6] Kastoris, D., Giotopoulos, K., and Papadopoulos, D. (2024). Neural network-based parameter estimation in dynamical systems, Information. 15(12). doi:10.3390/info15120809
[7] Kingma, D.P. and Ba, J. (2014). Adam: A method for stochastic optimization, arXiv preprint arXiv:1412.6980.
[8] Kumar, K. and Kostina, E. (2025). Machine learning in parameter estimation of nonlinear systems, European Physical Journal B. 98:60. doi:10.1140/epjb/s10051-025-00904-7
[9] Liang, R., Liu, W., Fu, Y., and Ma, M. (2024). Physics-informed deep learning for structural dynamics under moving load, International Journal of Mechanical Sciences. 284:109766. doi:10.1016/j.ijmecsci.2024.109766
[10] Luo, K., Zhao, J., Wang, Y., Li, J., Wen, J., Liang, J., Soekmadji, H., and Liao, S. (2025). Physics-informed neural networks for pde problems: A comprehensive review, Artificial Intelligence Review. 58(10):1--43.
[11] Ma, G., Wang, Z., Liu, W., Yang, Z., Huang, D., and Ding, H. (2025). Closed-loop parameter optimization for robotic machining using physics-informed machine learning and multiobjective optimization, IEEE Transactions on Automation Science and Engineering. 22:22410--22422. doi:10.1109/TASE.2025.3618242
[12] Nathasarma, R. and Roy, B.K. (2023). Physics-informed long-short-term memory neural network for parameters estimation of nonlinear systems, IEEE Transactions on Industry Applications. 59(5):5376--5384. doi:10.1109/TIA.2023.3280896
[13] Pintelon, R., Schoukens, J., Vandersteen, G., and Barbé, K. (2010). Estimation of nonparametric noise and frf models for multivariable systems—part i: Theory, Mechanical Systems and Signal Processing. 24(3):573--595. doi:10.1016/j.ymssp.2009.08.009
[14] Putkonen, A., Narsakka, J., Ranjan, G., Lindh, T., Sopanen, J., and Nevaranta, N. (2025). Commissioning of a modular active-magnetic-bearing-suspended rotor system, IEEE Open Journal of Industry Applications. 6:619--629. doi:10.1109/OJIA.2025.3596973
[15] Raissi, M., Perdikaris, P., and Karniadakis, G. (2019). Physics-informed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations, Journal of Computational Physics. 378:686--707. doi:10.1016/j.jcp.2018.10.045
[16] Rayhan, M., Abubakar, I., Ranjan, G., Sopanen, J., Lindh, T., and Nevaranta, N. (2025). Open access datasets for active magnetic bearing high-speed machinery, IEEE Access. 13:116491--116507. doi:10.1109/ACCESS.2025.3586026
[17] Rehtla, M., Abubakar, I., Putkonen, A., Shishkov, A., Nevaranta, N., and Lindh, T. (2024). Replicating existing axial magnetic bearing controller with a neural network, In 2024 Energy Conversion Congress & Expo Europe (ECCE Europe). pages 1--6. doi:10.1109/ECCEEurope62508.2024.10751915
[18] Simpkins, A. (2012). System identification: Theory for the user, 2nd edition (ljung, l, ; 1999) [on the sh.
[19] Song, C. and Alkhalifah, T.A. (2022). Wavefield reconstruction inversion via physics-informed neural networks, IEEE Transactions on Geoscience and Remote Sensing. 60:1--12. doi:10.1109/TGRS.2021.3123122
[20] Sun, J., Huang, Y., Yu, W., and Garcia-Ortiz, A. (2021). Nonlinear system identification: Prediction error method vs neural network, In 2021 10th International Conference on Modern Circuits and Systems Technologies (MOCAST). pages 1--4. doi:10.1109/MOCAST52088.2021.9493336
[21] Tutunji, T.A. (2016). Parametric system identification using neural networks, Applied Soft Computing. 47:251--261.
[22] Wang, D. and Zilouchian, A. (1998). Identification of discrete linear system in state space form using neural network, In Proceedings of the 1998 Second IEEE International Caracas Conference on Devices, Circuits and Systems. ICCDCS 98. On the 70th Anniversary of the MOSFET and 50th of the BJT. (Cat. No.98TH8350). pages 338--342. doi:10.1109/ICCDCS.1998.705860
[23] Wu, Y., Zhu, R., Cao, Z., Liu, Y., and Jiang, D. (2020). Model updating using frequency response functions based on sherman–morrison formula, Applied Sciences. 10(14). doi:10.3390/app10144985
[24] Zaman, A., Laila, D.S., and Birk, W. (2024). Towards empirical transfer function estimation for frequency domain closed-loop system identification, In 2024 10th International Conference on Control, Decision and Information Technologies (CoDIT). pages 946--951. doi:10.1109/CoDIT62066.2024.10708407
[25] Zhao, S., Peng, Y., Zhang, Y., and Wang, H. (2022). Parameter estimation of power electronic converters with physics-informed machine learning, IEEE Transactions on Power Electronics. 37(10):11567--11578. doi:10.1109/TPEL.2022.3176468
BibTeX:
@article{MIC-2026-2-1,
title={{Frequency-Domain Physics-Informed Neural Networks for Parameter Estimation}},
author={Tariq, Jawad and Lindh, Tuomo and Nevaranta, Niko},
journal={Modeling, Identification and Control},
volume={47},
number={2},
pages={39--50},
year={2026},
doi={10.4173/mic.2026.2.1},
publisher={Norwegian Society of Automatic Control}
};