pith. sign in

arxiv: 2604.01730 · v1 · submitted 2026-04-02 · 💻 cs.LG · cs.SY· eess.SY

Koopman-Based Nonlinear Identification and Adaptive Control of a Turbofan Engine

David Grasev This is my paper

Pith reviewed 2026-05-13 22:02 UTC · model grok-4.3

classification 💻 cs.LG cs.SYeess.SY
keywords Koopman operatorturbofan engineextended dynamic mode decompositionmodel predictive controladaptive controlfeedback linearizationnonlinear identificationengine pressure ratio
0
0 comments X p. Extension

The pith

A single Koopman model from meta-heuristic extended dynamic mode decomposition accurately predicts turbofan spool speeds and engine pressure ratio for reuse in multiple controllers.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper develops a physics-based component-level model of a two-spool turbofan engine to generate training data and then applies a meta-heuristic extended dynamic mode decomposition with a cost function focused on both spool speeds and EPR. This produces one time-varying Koopman model that supports two different controllers: an adaptive Koopman model predictive controller with disturbance observer and a Koopman-based feedback linearization controller used as benchmark. Evaluation under sea-level and varying flight conditions shows comparable steady-state performance but superior robustness for the adaptive MPC due to mismatch compensation, with EPR control also improving thrust response. A reader cares because the approach offers a reusable data-driven linear representation for nonlinear engine control without embedding the full physics model in real time.

Core claim

The meta-heuristic extended dynamic mode decomposition constructs a Koopman model that accurately predicts both spool speeds and EPR, enabling flexible reuse across control formulations; the adaptive Koopman MPC with disturbance observer exhibits superior robustness to varying flight conditions compared to the Koopman feedback linearization controller because it compensates for model mismatch.

What carries the argument

The time-varying Koopman model identified via meta-heuristic extended dynamic mode decomposition with a cost function designed to capture spool-speed dynamics and EPR; it provides a lifted linear representation of the nonlinear engine dynamics for controller design.

If this is right

  • The same identified model supports both spool-speed and EPR control configurations without retraining.
  • The adaptive Koopman MPC compensates for model mismatch and maintains performance when flight conditions change.
  • An EPR-focused control strategy produces faster thrust response than spool-speed-only strategies.
  • Koopman-based methods apply to multivariable nonlinear control of turbofan engines.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • If hardware tests confirm the results, the method could simplify real-time engine control software by avoiding full nonlinear models.
  • The approach may transfer to other nonlinear systems such as gas turbines or chemical processes where operating conditions vary.
  • Adding sensor noise or actuator limits to the validation would test whether the disturbance observer remains effective.

Load-bearing premise

The physics-based component-level model used to generate data accurately captures real turbofan engine dynamics so the identified Koopman model generalizes to untested flight conditions.

What would settle it

Large errors between the Koopman model's predicted spool speeds or EPR and measurements from actual engine runs under varying flight conditions would falsify the accuracy and reusability claim.

read the original abstract

This paper investigates Koopman operator-based approaches for multivariable control of a two-spool turbofan engine. A physics-based component-level model is developed to generate training data and validate the controllers. A meta-heuristic extended dynamic mode decomposition is developed, with a cost function designed to accurately capture both spool-speed dynamics and the engine pressure ratio (EPR), enabling the construction of a single Koopman model suitable for multiple control objectives. Using the identified time-varying Koopman model, two controllers are developed: an adaptive Koopman-based model predictive controller (AKMPC) with a disturbance observer and a Koopman-based feedback linearization controller (K-FBLC), which serves as a benchmark. The controllers are evaluated for two control strategies, namely configurations of spool speeds and EPR, under both sea-level and varying flight conditions. The results demonstrate that the proposed identification approach enables accurate predictions of both spool speeds and EPR, allowing the Koopman model to be reused flexibly across different control formulations. While both control strategies achieve comparable performance in steady conditions, the AKMPC exhibits superior robustness compared with the K-FBLC under varying flight conditions due to its ability to compensate for model mismatch. Moreover, the EPR control strategy improves the thrust response. The study highlights the applicability of Koopman-based control and demonstrates the advantages of the AKMPC-based framework for robust turbofan engine control.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

3 major / 3 minor

Summary. The paper develops a physics-based component-level model of a two-spool turbofan engine to generate training data, then applies a meta-heuristic extended dynamic mode decomposition (EDMD) to identify a single Koopman operator model that predicts both spool speeds and engine pressure ratio (EPR). This model is used to design an adaptive Koopman-based model predictive controller (AKMPC) with disturbance observer and a Koopman-based feedback linearization controller (K-FBLC) as benchmark. Closed-loop simulations under sea-level static and varying flight conditions compare spool-speed and EPR control strategies, claiming accurate predictions, flexible model reuse, and superior robustness of AKMPC to model mismatch.

Significance. If the central claims hold under stronger validation, the work would demonstrate practical utility of Koopman methods for multivariable nonlinear control of turbofan engines, particularly the reuse of one identified model across objectives and the adaptive compensation mechanism in AKMPC. This could inform data-driven designs where flight-condition variation and model uncertainty are present, provided the simulation results translate beyond the training environment.

major comments (3)
  1. [Abstract and §5] Abstract and §5: The claim that AKMPC exhibits superior robustness to model mismatch under varying flight conditions is not supported by the experimental design, because data generation, Koopman identification, and all closed-loop evaluations occur inside the identical physics-based simulator; artificial mismatch injected within this loop cannot establish generalization to real turbofan dynamics such as sensor noise or unmodeled aerothermodynamic effects.
  2. [§3] §3: The meta-heuristic EDMD procedure is central to the identification claim, yet no ablation study, comparison against standard EDMD, or sensitivity analysis of the cost-function weights on spool-speed versus EPR accuracy is provided; without these, it is unclear whether the reported prediction performance stems from the meta-heuristic or from the underlying data.
  3. [§4 and §5] §4 and §5: No quantitative prediction metrics (RMSE, maximum absolute error, or cross-validation error on hold-out trajectories) or validation protocol details are reported for the identified Koopman model, which directly undermines the assertion that the model enables “accurate predictions” reusable across control formulations.
minor comments (3)
  1. [§2] §2: Notation for the lifted-state Koopman operator and the time-varying A/B matrices is introduced without an explicit comparison table to standard DMD, which could confuse readers familiar with the literature.
  2. [Figure 4 and Figure 6] Figure 4 and Figure 6: The time-response plots for varying flight conditions lack shaded uncertainty bands or tabulated steady-state error statistics, reducing clarity of the robustness comparison.
  3. [References] References: Several recent works on Koopman-based aerospace control (e.g., applications to aircraft or gas-turbine systems) are not cited, which would help situate the contribution.

Simulated Author's Rebuttal

3 responses · 1 unresolved

We thank the referee for the constructive comments. We address each major point below and outline revisions to improve clarity, rigor, and substantiation of the claims.

read point-by-point responses

  1. Referee: [Abstract and §5] Abstract and §5: The claim that AKMPC exhibits superior robustness to model mismatch under varying flight conditions is not supported by the experimental design, because data generation, Koopman identification, and all closed-loop evaluations occur inside the identical physics-based simulator; artificial mismatch injected within this loop cannot establish generalization to real turbofan dynamics such as sensor noise or unmodeled aerothermodynamic effects.

    Authors: We agree that the entire study, including data generation and controller evaluation, is performed within the same high-fidelity physics-based simulator. The injected mismatches are intended to represent parametric uncertainties and flight-condition variations, but we acknowledge this does not constitute proof of generalization to physical hardware or unmodeled effects such as sensor noise. We will revise the abstract and §5 to explicitly state that the robustness comparison holds under simulated model mismatch and varying conditions within the simulator framework. We will also add a dedicated limitations paragraph discussing the simulation-only nature of the results and the need for future hardware validation. revision: partial

  2. Referee: [§3] §3: The meta-heuristic EDMD procedure is central to the identification claim, yet no ablation study, comparison against standard EDMD, or sensitivity analysis of the cost-function weights on spool-speed versus EPR accuracy is provided; without these, it is unclear whether the reported prediction performance stems from the meta-heuristic or from the underlying data.

    Authors: We will add an ablation study in §3 that compares the meta-heuristic EDMD against standard EDMD on the same dataset. We will also include a sensitivity analysis varying the cost-function weights for spool-speed versus EPR terms and report the resulting prediction accuracies. These additions will clarify the contribution of the meta-heuristic optimization. revision: yes

  3. Referee: [§4 and §5] §4 and §5: No quantitative prediction metrics (RMSE, maximum absolute error, or cross-validation error on hold-out trajectories) or validation protocol details are reported for the identified Koopman model, which directly undermines the assertion that the model enables “accurate predictions” reusable across control formulations.

    Authors: We will include quantitative metrics (RMSE, maximum absolute error, and cross-validation error on hold-out trajectories) for the Koopman model predictions of both spool speeds and EPR. We will also detail the validation protocol, including data partitioning and evaluation procedures, in §4. These quantitative results will be referenced in §5 to support the claims of accurate predictions and model reusability. revision: yes

standing simulated objections not resolved
  • Real-world validation on physical turbofan engine hardware or flight test data, as the study is limited to simulation using a physics-based component-level model.

Circularity Check

0 steps flagged

No significant circularity; derivation remains self-contained

full rationale

The paper first constructs an independent physics-based component-level turbofan model to generate training trajectories, then applies meta-heuristic EDMD (with an explicitly designed cost function penalizing spool-speed and EPR errors) to obtain a Koopman approximation, and finally designs and closes the loop with AKMPC and K-FBLC controllers whose performance is measured against the same simulator. None of these steps collapses by construction: the identification objective is dynamics matching, not the final tracking metrics; the reported robustness advantage is a comparative outcome between two distinct controller architectures on the same plant; and no self-citation, uniqueness theorem, or ansatz is invoked to force the central claims. Validation inside the training simulator is standard engineering practice and does not equate the reported controller performance to the identification inputs.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Only the abstract is available; no explicit free parameters, axioms, or invented entities are stated. The approach relies on standard Koopman operator theory and optimization methods whose details are not provided.

pith-pipeline@v0.9.0 · 5541 in / 1338 out tokens · 47312 ms · 2026-05-13T22:02:14.672090+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

57 extracted references · 57 canonical work pages

  1. [1]

    American Institute for Aeronautics and Astronau- tics, Reston, Virginia (2016)

    Mattingly, J.D., Boyer, K.M.: Elements of Propul- sion: Gas Turbines and Rockets, Second Edition. American Institute for Aeronautics and Astronau- tics, Reston, Virginia (2016)

    work page 2016

  2. [2]

    Amer Inst Aero & Astro, Reston, Virginia (2009)

    Jaw, L.C., Mattingly, J.D.: Aircraft Engine Con- trols: Design, System Analysis, And Health Moni- toring. Amer Inst Aero & Astro, Reston, Virginia (2009)

    work page 2009

  3. [3]

    Journal of Aerospace Engineering26(2), 422–438 (2013) https://doi.org/ 10.1061/(ASCE)AS.1943-5525.0000296

    Garg, S.: Aircraft turbine engine control research at nasa glenn research center. Journal of Aerospace Engineering26(2), 422–438 (2013) https://doi.org/ 10.1061/(ASCE)AS.1943-5525.0000296

    work page doi:10.1061/(asce)as.1943-5525.0000296 2013

  4. [4]

    Propulsion and Power Research11(1), 1–57 (2022) https://doi.org/10.1016/j.jppr.2022.02.003

    Lv, C., Chang, J., Bao, W., Yu, D.: Recent research progress on airbreathing aero-engine control algo- rithm. Propulsion and Power Research11(1), 1–57 (2022) https://doi.org/10.1016/j.jppr.2022.02.003

    work page doi:10.1016/j.jppr.2022.02.003 2022

  5. [5]

    Cheng, D., Liu, L., Yu, Z.: A nonlinearH ∞ set-point control method for turbofan engines with distur- bances. Int. J. Control Autom. Syst.16, 3062–3074 (2021) https://doi.org/10.1007/s12555-020-0436-3

    work page doi:10.1007/s12555-020-0436-3 2021

  6. [6]

    In: 1989 American Control Conference, pp

    Garg, S.: Turbofan engine control system design using the lqg/ltr methodology. In: 1989 American Control Conference, pp. 134–141 (1989). https:// doi.org/10.23919/ACC.1989.4790178

    work page doi:10.23919/acc.1989.4790178 1989

  7. [8]

    Chinese Journal of Aeronautics35(6), 125–136 (2022) https://doi.org/10.1016/j.cja.2021.09.018

    Zhu, Y., Huang, J., Pan, M., Zhou, W.: Direct thrust control for multivariable turbofan engine based on affine linear parameter-varying approach. Chinese Journal of Aeronautics35(6), 125–136 (2022) https://doi.org/10.1016/j.cja.2021.09.018

    work page doi:10.1016/j.cja.2021.09.018 2022

  8. [9]

    Aerospace Science and Technology157, 109754 (2025) https://doi.org/10.1016/j.ast.2024.109754

    Li, S., Wang, Y., Zhang, H.: A multivariable adap- tive control method for aeroengine with hinf per- formance considering engine output limitation pro- tection based on fully adjustable neural network. Aerospace Science and Technology157, 109754 (2025) https://doi.org/10.1016/j.ast.2024.109754

    work page doi:10.1016/j.ast.2024.109754 2025

  9. [10]

    Chinese Journal of Aeronautics33(6), 1756–1773 (2020) https://doi.org/10.1016/j.cja.2020.01.011

    Pang, S., Li, Q., Zhang, H.: A new online mod- elling method for aircraft engine state space model. Chinese Journal of Aeronautics33(6), 1756–1773 (2020) https://doi.org/10.1016/j.cja.2020.01.011

    work page doi:10.1016/j.cja.2020.01.011 2020

  10. [11]

    Aerospace Science and Technology105, 105951 (2020) https://doi.org/10.1016/j.ast.2020.105951

    Pang, S., Li, Q., Feng, H.: A hybrid onboard adap- tive model for aero-engine parameter prediction. Aerospace Science and Technology105, 105951 (2020) https://doi.org/10.1016/j.ast.2020.105951

    work page doi:10.1016/j.ast.2020.105951 2020

  11. [12]

    Chi- nese Journal of Aeronautics34(12), 57–72 (2021) https://doi.org/10.1016/j.cja.2021.04.002

    Pang, S., Jafari, S., Nikolaidis, T., Li, Q.: A novel model-based multivariable framework for aircraft gas turbine engine limit protection control. Chi- nese Journal of Aeronautics34(12), 57–72 (2021) https://doi.org/10.1016/j.cja.2021.04.002

    work page doi:10.1016/j.cja.2021.04.002 2021

  12. [13]

    Aerospace Science and Technology158, 109927 (2025) https://doi.org/10.1016/j.ast.2025

    Yao, J., Li, R., Guo, Y., Wu, H., Cui, G., Zhou, Z.: Online performance seeking control of aero- engine based on stabilization-optimization dual loop. Aerospace Science and Technology158, 109927 (2025) https://doi.org/10.1016/j.ast.2025. 109927

    work page doi:10.1016/j.ast.2025 2025

  13. [14]

    Nonlinear Dynamics114(3), 205 (2026) https://doi.org/10.1007/s11071-025-12070-7

    Grasev, D.: Koopman eigenfunction-based identi- fication and optimal nonlinear control of turbojet engine. Nonlinear Dynamics114(3), 205 (2026) https://doi.org/10.1007/s11071-025-12070-7

    work page doi:10.1007/s11071-025-12070-7 2026

  14. [15]

    Aerospace Science and Technology138, 108319 (2023) https://doi.org/10

    Liu, Z., Huang, Y., Gou, L., Fan, D.: A robust adaptive linear parameter-varying gain-scheduling 18 controller for aeroengines. Aerospace Science and Technology138, 108319 (2023) https://doi.org/10. 1016/j.ast.2023.108319

    work page arXiv 2023

  15. [16]

    IEEE Transactions on Aerospace and Electronic Systems59(6), 7440–7454 (2023) https://doi.org/ 10.1109/TAES.2023.3288854

    Chen, Q., Sheng, H., Li, J., Liu, T.: Model-based improved advanced adaptive performance recovery control method for a commercial turbofan engine. IEEE Transactions on Aerospace and Electronic Systems59(6), 7440–7454 (2023) https://doi.org/ 10.1109/TAES.2023.3288854

    work page doi:10.1109/taes.2023.3288854 2023

  16. [18]

    A gaussian mixture distribution-based adaptive sampling method for physics-informed neural networks.Engineering Applications of Artificial Intelligence, 135:108770, 2024

    Wei, Z., Jafari, S., Zhang, S., Nikolaidis, T.: Hybrid wiener model: An on-board approach using post-flight data for gas turbine aero- engines modelling. Applied Thermal Engineer- ing184, 116350 (2021) https://doi.org/10.1016/j. applthermaleng.2020.116350

    work page doi:10.1016/j 2021

  17. [19]

    Energy242, 123030 (2022) https://doi.org/10.1016/j.energy.2021.123030

    Wei, Z., Zhang, S., Jafari, S., Nikolaidis, T.: Self- enhancing model-based control for active transient protection and thrust response improvement of gas turbine aero-engines. Energy242, 123030 (2022) https://doi.org/10.1016/j.energy.2021.123030

    work page doi:10.1016/j.energy.2021.123030 2022

  18. [20]

    Yang, W., Tang, Y., Zhou, W., Yang, G., Huang, J., Cui, T.: Off-equilibrium linearization-based control of nonlinear time-delay system and application to a turbofan engine. ASME J. Eng. Gas Turbines Power 146(10), 101017 (2024) https://doi.org/10.1115/1. 4065993

    work page doi:10.1115/1 2024

  19. [21]

    Aerospace Science and Technology126, 107586 (2022) https://doi.org/10

    Singh, R., Maity, A., Nataraj, P.S.V.: Dynamic modeling and robust nonlinear control of a labo- ratory gas turbine engine. Aerospace Science and Technology126, 107586 (2022) https://doi.org/10. 1016/j.ast.2022.107586

    work page arXiv 2022

  20. [22]

    Energy217, 119363 (2021) https://doi.org/10

    Palmieri, A., Lanzarotto, D., Cacciacarne, S., Torre, I., Bonfiglio, A.: An innovative sliding mode load controller for gas turbine power generators: Design and experimental validation via real-time simula- tion. Energy217, 119363 (2021) https://doi.org/10. 1016/j.energy.2020.119363

    work page arXiv 2021

  21. [23]

    Energy121, 491–512 (2017) https://doi.org/10

    Bonfiglio, A., Cacciacarne, S., Invernizzi, M., Proco- pio, R., Schiano, S., Torre, I.: Gas turbine generating units control via feedback linearization approach. Energy121, 491–512 (2017) https://doi.org/10. 1016/j.energy.2017.01.048

    work page 2017

  22. [24]

    IEEE Robotics and Automation Letters5(2), 2070–2077 (2020) https://doi.org/10.1109/LRA.2020.2970572

    L’Erario, G., Fiorio, L., Nava, G., Bergonti, F., Mohamed, H.A.O., Benenati, E., Traversaro, S., Pucci, D.: Modeling, identification and control of model jet engines for jet powered robotics. IEEE Robotics and Automation Letters5(2), 2070–2077 (2020) https://doi.org/10.1109/LRA.2020.2970572

    work page doi:10.1109/lra.2020.2970572 2070

  23. [25]

    Turbo Expo, vol

    Brunell, B.J., Viassolo, D.E., Prasanth, R.: Model Adaptation and Nonlinear Model Predictive Control of an Aircraft Engine. Turbo Expo, vol. Volume 2: Turbo Expo 2004, pp. 673–682 (2004). https://doi. org/10.1115/GT2004-53780

    work page doi:10.1115/gt2004-53780 2004

  24. [26]

    Aerospace Science and Tech- nology92, 99–113 (2019) https://doi.org/10.1016/ j.ast.2019.05.061

    Montazeri-Gh, M., Rasti, A., Jafari, A., Ehteshami, M.: Design and implementation of mpc for turbofan engine control system. Aerospace Science and Tech- nology92, 99–113 (2019) https://doi.org/10.1016/ j.ast.2019.05.061

    work page 2019

  25. [27]

    A novel tensor-based modal decomposition method for reduced order modeling and optimal sparse sensor placement,

    Song, J., Chen, Y., Li, W., Wang, Y., Zhang, H.: Nonlinear model predictive control method for high- speed helicopter power system based on integrated onboard model. Aerospace Science and Technology 161, 110093 (2025) https://doi.org/10.1016/j.ast. 2025.110093

    work page doi:10.1016/j.ast 2025

  26. [28]

    International Journal of Aeronautical and Space Sciences20, 933–939 (2019) https://doi.org/ 10.1007/s42405-019-00191-4

    Zheng, Q., Pang, S., Zhang, H., Hu, Z.: A study on aero-engine direct thrust control with nonlin- ear model predictive control based on deep neural network. International Journal of Aeronautical and Space Sciences20, 933–939 (2019) https://doi.org/ 10.1007/s42405-019-00191-4

    work page doi:10.1007/s42405-019-00191-4 2019

  27. [29]

    Chinese Journal of Aeronautics35(4), 66–81 (2022) https://doi.org/10.1016/j.cja.2021.08.024

    Pang, S., Jafari, S., Nikolaidis, T., Li, Q.: Reduced- dimensional mpc controller for direct thrust control. Chinese Journal of Aeronautics35(4), 66–81 (2022) https://doi.org/10.1016/j.cja.2021.08.024

    work page doi:10.1016/j.cja.2021.08.024 2022

  28. [30]

    Aerospace Science and Tech- nology118, 106983 (2021) https://doi.org/10.1016/ j.ast.2021.106983

    Pang, S., Li, Q., Ni, B.: Improved nonlinear mpc for aircraft gas turbine engine based on semi-alternative optimization strategy. Aerospace Science and Tech- nology118, 106983 (2021) https://doi.org/10.1016/ j.ast.2021.106983

    work page arXiv 2021

  29. [31]

    Energy230, 120700 (2021) https://doi.org/10.1016/j.energy.2021.120700

    Ibrahem, I.M.A., Akhrif, O., Moustapha, H., Staniszewski, M.: Nonlinear general- ized predictive controller based on ensemble of narx models for industrial gas tur- bine engine. Energy230, 120700 (2021) https://doi.org/10.1016/j.energy.2021.120700

    work page doi:10.1016/j.energy.2021.120700 2021

  30. [32]

    Chinese Journal of Aeronautics 35(12), 59–71 (2022) https://doi.org/10.1016/j.cja

    Wang, Y., Huang, J., Zhou, W., Lu, F., Xu, W.: Neural network-based model predictive control with fuzzy-sqp optimization for direct thrust control of turbofan engine. Chinese Journal of Aeronautics 35(12), 59–71 (2022) https://doi.org/10.1016/j.cja. 2022.04.012

    work page doi:10.1016/j.cja 2022

  31. [33]

    Applied Sciences12(10) (2022) https://doi.org/10.3390/app12104874

    Ji, R., Huang, X., Zhao, X.: Active jet noise con- trol of turbofan engine based on explicit model predictive control. Applied Sciences12(10) (2022) https://doi.org/10.3390/app12104874

    work page doi:10.3390/app12104874 2022

  32. [34]

    The design of stretch: A compact, lightweight mobile manipulator for indoor human environments,

    Momin, A.J.A., Nava, G., L’Erario, G., Mohamed, 19 H.A.O., Bergonti, F., Vanteddu, P.R., Braghin, F., Pucci, D.: Nonlinear model identification and observer design for thrust estimation of small-scale turbojet engines. In: 2022 International Conference on Robotics and Automation (ICRA), pp. 5879– 5885 (2022). https://doi.org/10.1109/ICRA46639. 2022.9812283

    work page doi:10.1109/icra46639 2022

  33. [35]

    (eds.): The Koopman Operator in Systems and Control

    Mauroy, A., Mezi´ c, I., Susuki, Y. (eds.): The Koopman Operator in Systems and Control. Springer, Cham, Switzerland (2020). https://doi. org/10.1007/978-3-030-35713-9

    work page doi:10.1007/978-3-030-35713-9 2020

  34. [36]

    IFAC-PapersOnLine49(18), 716–723 (2016) https://doi.org/10.1016/j.ifacol.2016.10.250

    Surana, A., Banaszuk, A.: Linear observer synthe- sis for nonlinear systems using koopman operator framework. IFAC-PapersOnLine49(18), 716–723 (2016) https://doi.org/10.1016/j.ifacol.2016.10.250 . 10th IFAC Symposium on Nonlinear Control Sys- tems NOLCOS 2016

    work page doi:10.1016/j.ifacol.2016.10.250 2016

  35. [37]

    Journal of Non- linear Science30(5), 1973–2006 (2020) https://doi

    Surana, A.: Koopman Operator Framework for Time Series Modeling and Analysis. Journal of Non- linear Science30(5), 1973–2006 (2020) https://doi. org/10.1007/s00332-017-9441-y

    work page doi:10.1007/s00332-017-9441-y 1973

  36. [38]

    SIAM Journal on Applied Dynamical Sys- tems17(1), 909–930 (2018) https://doi.org/10

    Proctor, J.L., Brunton, S.L., Kutz, J.N.: Gener- alizing koopman theory to allow for inputs and control. SIAM Journal on Applied Dynamical Sys- tems17(1), 909–930 (2018) https://doi.org/10. 1137/16M1062296

    work page 2018

  37. [39]

    Machine Learning: Science and Technology2(3), 035023 (2021) https://doi.org/10.1088/2632-2153/ abf0f5

    Kaiser, E., Kutz, J.N., Brunton, S.L.: Data-driven discovery of koopman eigenfunctions for control. Machine Learning: Science and Technology2(3), 035023 (2021) https://doi.org/10.1088/2632-2153/ abf0f5

    work page doi:10.1088/2632-2153/ 2021

  38. [40]

    Korda, I

    Korda, M., Mezi´ c, I.: Linear predictors for nonlinear dynamical systems: Koopman operator meets model predictive control. Automatica93, 149–160 (2018) https://doi.org/10.1016/j.automatica.2018.03.046

    work page doi:10.1016/j.automatica.2018.03.046 2018

  39. [41]

    IEEE Transactions on Automatic Control 65(12), 5114–5129 (2020) https://doi.org/10.1109/ TAC.2020.2978039

    Korda, M., Mezi´ c, I.: Optimal construction of koopman eigenfunctions for prediction and con- trol. IEEE Transactions on Automatic Control 65(12), 5114–5129 (2020) https://doi.org/10.1109/ TAC.2020.2978039

    work page arXiv 2020

  40. [42]

    Mechatronics86, 102871 (2022) https://doi.org/10.1016/j.mechatronics

    Chen, J., Dang, Y., Han, J.: Offset-free model predictive control of a soft manipulator using the koopman operator. Mechatronics86, 102871 (2022) https://doi.org/10.1016/j.mechatronics. 2022.102871

    work page doi:10.1016/j.mechatronics 2022

  41. [43]

    The International Journal of Robotics Research44(13), 1–27 (2025) https://doi

    Singh, R., Sah, C.K., Keshavan, J.: Adaptive koop- man embedding for robust control of nonlinear dynamical systems. The International Journal of Robotics Research44(13), 1–27 (2025) https://doi. org/10.1177/02783649251341907

    work page doi:10.1177/02783649251341907 2025

  42. [44]

    IEEE Access 13, 13944–13958 (2025) https://doi.org/10.1109/ ACCESS.2025.3530984

    Kim, J.S., Quan, Y.S., Chung, C.C., Choi, W.Y.: K-smpc: Koopman operator-based stochas- tic model predictive control for enhanced lat- eral control of autonomous vehicles. IEEE Access 13, 13944–13958 (2025) https://doi.org/10.1109/ ACCESS.2025.3530984

    work page arXiv 2025

  43. [45]

    Applied Energy381, 125117 (2025) https:// doi.org/10.1016/j.apenergy.2024.125117

    Zhou, J., Jia, Y., Sun, C.: Flywheel energy storage system controlled using tube-based deep koopman model predictive control for wind power smooth- ing. Applied Energy381, 125117 (2025) https:// doi.org/10.1016/j.apenergy.2024.125117

    work page doi:10.1016/j.apenergy.2024.125117 2025

  44. [46]

    GasTurb GmbH, Aachen, Germany (2026)

    Kurzke, J.: GasTurb 15: Design and Off-Design Performance of Gas Turbines. GasTurb GmbH, Aachen, Germany (2026). GasTurb GmbH. Avail- able at https://www.gasturb.com/Downloads/ Manuals/GasTurb15.pdf

    work page 2026

  45. [47]

    In: 2024 21st Interna- tional Conference on Mechatronics - Mechatronika (ME), pp

    Grasev, D., J´ ılek, A.: A novel approach to modeling of compressor characteristics. In: 2024 21st Interna- tional Conference on Mechatronics - Mechatronika (ME), pp. 158–162 (2024). https://doi.org/10.1109/ ME61309.2024.10789753

    work page arXiv 2024

  46. [48]

    Turbo Expo, vol

    Kurzke, J.: How to Get Component Maps for Air- craft Gas Turbine Performance Calculations. Turbo Expo, vol. Volume 5: Manufacturing Materials and Metallurgy; Ceramics; Structures and Dynamics; Controls, Diagnostics and Instrumentation; Educa- tion; General, pp. 005–16001 (1996). https://doi. org/10.1115/96-GT-164

    work page doi:10.1115/96-gt-164 1996

  47. [49]

    Applied Thermal Engineering236, 121523 (2024) https://doi.org/10.1016/j.applthermaleng.2023

    Yang, Y., Nikolaidis, T., Jafari, S., Pilidis, P.: Gas turbine engine transient performance and heat transfer effect modelling: A comprehensive review, research challenges, and exploring the future. Applied Thermal Engineering236, 121523 (2024) https://doi.org/10.1016/j.applthermaleng.2023. 121523

    work page doi:10.1016/j.applthermaleng.2023 2024

  48. [50]

    Annual Review of Fluid Mechanics45(Volume 45, 2013), 357–378 (2013) https://doi.org/10.1146/ annurev-fluid-011212-140652

    Mezi´ c, I.: Analysis of fluid flows via spectral properties of the koopman operator. Annual Review of Fluid Mechanics45(Volume 45, 2013), 357–378 (2013) https://doi.org/10.1146/ annurev-fluid-011212-140652

    work page 2013

  49. [51]

    Physica D: Non- linear Phenomena406, 132416 (2020) https://doi

    Klus, S., N¨ uske, F., Peitz, S., Niemann, J.-H., Clementi, C., Sch¨ utte, C.: Data-driven approxima- tion of the koopman generator: Model reduction, system identification, and control. Physica D: Non- linear Phenomena406, 132416 (2020) https://doi. org/10.1016/j.physd.2020.132416

    work page doi:10.1016/j.physd.2020.132416 2020

  50. [52]

    IEEE Access13(1), 91972–91988 (2025) https://doi.org/10.1109/ACCESS.2025.3573472 20

    Grasev, D.: Data-driven identification of gas tur- bine engine dynamics via koopman operator genetic algorithm. IEEE Access13(1), 91972–91988 (2025) https://doi.org/10.1109/ACCESS.2025.3573472 20

    work page doi:10.1109/access.2025.3573472 2025

  51. [53]

    Korˇ ze and A

    Chen, Z., Chen, X., Liu, J., Cen, L., Gui, W.: Learn- ing model predictive control of nonlinear systems with time-varying parameters using koopman oper- ator. Applied Mathematics and Computation470, 128577 (2024) https://doi.org/10.1016/j.amc.2024. 128577

    work page doi:10.1016/j.amc.2024 2024

  52. [54]

    Dover Publi- cations, Mineola, NY (1987)

    Friedland, B.: Control System Design. Dover Publi- cations, Mineola, NY (1987)

    work page 1987

  53. [55]

    Prentice Hall, Michigan (1991)

    Slotine, J.J.E., Li, W.: Applied Nonlinear Control. Prentice Hall, Michigan (1991)

    work page 1991

  54. [56]

    In: Proceedings of ICNN’95 - International Conference on Neural Networks, vol

    Kennedy, J., Eberhart, R.: Particle swarm optimiza- tion. In: Proceedings of ICNN’95 - International Conference on Neural Networks, vol. 4, pp. 1942– 1948 (1995). https://doi.org/10.1109/ICNN.1995. 488968

    work page doi:10.1109/icnn.1995 1942

  55. [57]

    Morgan Kaufmann, San Francisco, CA (2001)

    Eberhart, R.C., Shi, Y., Kennedy, J.: Swarm Intel- ligence, 1st edn. Morgan Kaufmann, San Francisco, CA (2001)

    work page 2001

  56. [58]

    Nelder, J.A., Mead, R.: A simplex method for func- tion minimization. Comput. J.7, 308–313 (1965) https://doi.org/10.1093/COMJNL/7.4.308

    work page doi:10.1093/comjnl/7.4.308 1965

  57. [59]

    Springer, Cham (2020)

    Volponi, A.J.: Gas Turbine Parameter Corrections. Springer, Cham (2020). https://doi.org/10.1007/ 978-3-030-41076-6 21

    work page 2020