“Investigation of Fluid Power Systems Modelling Using Lagrangian Neural Network Structures”
Authors: Jeppe H. Andersen, Jokin L. Berakoetxea, Esben Kaasgaard Lillie Lehmann, Stas Kokotovic and Per Johansen,Affiliation: Aalborg University
Reference: 2026, Vol 47, No 1, pp. 23-38.
Keywords: Neural Networks, LNN, Internal Energy, Continuity
Abstract: Modelling Fluid Power Systems is commonly done using first principle models which can be computationally demanding or time-consuming to fit. This paper introduces and explores the use of three Neural Network Structures in Fluid Power Systems modelling, referred to as Pressure Derivative Neural Network Structure, Internal Energy Neural Network Structure and Continuity Based Neural Network Structure. The structures are Lagrangian Neural Network like structures, which can parametrise arbitrary Lagrangians using Neural Networks. Lagrangian Neural Networks have predominantly been applied in mechanical systems, with no prior applications in Fluid Power Systems. This approach is applied to a valve-controlled cylinder drive as a case study. Datasets are both generated from a simulation and a lab setup. Primarily, a training scheme comparing states integrated ahead in time, using NN-predicted derivatives, are investigated. The results show promising performance training on simulation data, also including noise. However, training on the lab data proved to inhibit the Neural Network Structures from learning the dynamics. An investigation into the differences between simulation and lab data revealed that the discrepancies affected the Internal Energy Neural Network Structure and Continuity Based Neural Network Structure differently. Both structures were most affected by inaccurate valve position and velocity data.
PDF (2042 Kb)
DOI: 10.4173/mic.2026.1.3References:
[1] Book, W.J. (1983). Recursive Lagrangian Dynamics of Flexible Manipulator Arms via Transformation Matrices, Carnegie-Mellon University.
[2] Brunton, S.L. and Kutz, J.N. (2019). Data-Driven Science and Engineering: Machine Learning, Dynamical Systems, and Control, Cambridge University Press, USA, 1st edition.
[3] Cranmer, M. (2020). Lagrangian Neural Networks, doi:10.48550/arXiv.2003.04630
[4] Hansen, A.H. (2023). Fluid Power Systems, Number ISBN: 978-3-031-15089-0 in eBook. Springer Cham.
[5] M.Raissi, G. E.K., P.P. (2017). Inferring solutions of differential equations using noisy multi-fidelity data, 2017, doi:10.1016/j.jcp.2017.01.060
[6] M.Raissi, G. E.K., P.P. (2017). Machine learning of linear differential equations using Gaussian processes, 2017, doi:10.1016/j.jcp.2017.07.050
[7] M.Raissi, G. E.K., P.P. (2019). Physics-informed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations, doi:10.1016/j.jcp.2018.10.045
[8] MathWorks. (2024). PRBS Input Signals, Last checked: 22-10-24.
[9] MichaelLutter, C. R. J.P. (2019). Deep Lagrangian Networks: Using Physics as Model Prior for Deep Learning, doi:10.48550/arXiv.1907.04490
[10] Moog. (2020). Direct Drive Analog Control Servo Valves D633 and D634, Last checked: 04-11-24.
[11] Sundararajan, D. (2022). Control Systems: An Introduction, Springer Nature.
BibTeX:
@article{MIC-2026-1-3,
title={{Investigation of Fluid Power Systems Modelling Using Lagrangian Neural Network Structures}},
author={Andersen, Jeppe H. and Berakoetxea, Jokin L. and Lehmann, Esben Kaasgaard Lillie and Kokotovic, Stas and Johansen, Per},
journal={Modeling, Identification and Control},
volume={47},
number={1},
pages={23--38},
year={2026},
doi={10.4173/mic.2026.1.3},
publisher={Norwegian Society of Automatic Control}
};