The GTE Turbine Thermal State Assessment Employing Neural Networks

Aeronautical and Space-Rocket Engineering


Аuthors

Grigor’ev E. M.*, Falaleev S. V.**

Samara National Research University named after Academician S.P. Korolev, 34, Moskovskoye shosse, Samara, 443086, Russia

*e-mail: grigory1287@gmail.com
**e-mail: falaleev.sv@ssau.ru

Abstract

When designing an aircraft engine, as well as its workability analyzing while its operation in transient conditions, thermal computations performing of its structure is necessary. Computational method employing full-scale thermo-mechanical model is laborious and time-consuming. The authors propose a structure thermal state predicting technique at the engine work process parameters variation by creating a simplified thermal model and neural networks application, and transfer learning on the example of a micro gas turbine engine turbine (micro-GTE). The said technique requires a large number of finite element computations of the thermal state of the turbine parts in MATLAB employing various combinations of boundary conditions, as well as limited set of experimental data.

In the course of the studies, various solutions for the model clarification, such as more denser Biot numbers distribution, parameters changing of the last hidden layer for transfer learning and experimental data set limiting, were tried out. The results of testing isolated from each other methods for the neuron network operation modification revealed that restriction of the experimental data set size, achieved by the data set division by the types of maneuvers, was most effective. The results of testing isolated from each other methods for the neuron network operation refining revealed that restriction of the experimental data set size, achieved by the data set division by the types of maneuvers, was most effective. After the process optimization, the result of learning is more closer to the experimental data.

This inference indicates the possibility of improving the results by obtaining the experimental data with lower noise and greater diversity of maneuvers. The extra data such as heat transfer coefficients and temperature near the surfaces non-contacting with the main gas flow, as well as general conditions of the gas turbine unit operation may be handy for the results accuracy improving. All that may help more accurate finite element modeling of non-stationary thermal process in the gas turbine structure. There is a possibility as well of considering more complex structures of the thermal machines assemblages for obtaining more accurate digital simulation results of non-stationary thermal processes.

Keywords:

gas turbine engine, multilayer perceptron, transfer learning, thermal map, finite element model, non-stationary thermal process

References

  1. Ehzrokhi Yu.A., Kalenskii S.M. Trudy MAI, 2022, no. 122. DOI: 10.34759/trd-2022-122-19

  2. Ezrokhi Yu.A., Khoreva E.A. Estimation of inlet airflow non-uniformity effect on turbofan thrust. Aerospace MAI Journal, 2018, vol. 25, no. 2, pp. 99-108.

  3. Klinskii B.M. Studying the flow non-uniformity impact at the inlet on the aircraft gas turbine engine basic parameters under the simulated altitude-speed conditions. Aerospace MAI Journal, 2023, vol. 30, no. 1, pp. 117-130. DOI: 10.34759/vst-2023-1-117-130

  4. Shevchenko I.V., Rogalev A.N., Rogalev N.D., Komarov I.I., Bryzgunov P.A. Experimental study of heat transfer in slotted channels of gas turbine engines cooled blades with modified pin heat transfer intensifiers. Aerospace MAI Journal, 2023, vol. 30, no. 3, pp. 92-100.

  5. Komp'yuternye modeli i modelirovanie. Tsifrovye dvoiniki izdelii. Obshchie polozheniya. GOST R 57700.37–2021 (Computer models and simulation. Digital twins of products. General provisions, State Standard R 57700.37–2021), Moscow, Rossiiskii institut standartizatsii, 2021, 15 p.

  6. Prokhorov A., Lysachev M.  Tsifrovoi dvoinik. Analiz, trendy, mirovoi opyt (Digital double. Analysis, trends, world experience). Moscow, Al'yanSPrint, 2020, 401 p.

  7. Falaleev S.V., Zrelov V.A., Shchemelev V.I. Nasosy. Turbiny. Sistemy, 2020, no. 1(34), pp. 39-45.

  8. Grigor'ev E.M., Badykov R.R., Falaleev S.V. Materialy Mezhdunarodnoi molodezhnoi nauchnoi konferentsii, posvyashchennoi 60-letiyu poleta v kosmos YU.A. Gagarina “XVI Korolevskie chteniya” (5-7 October 2021; Samara). Samara, SSAU, 2021, pp. 207-209.

  9. Baumann M., Koch C., Staudacher S. Experimental identification of steady-state turbomachinery heat transfer using nondimensional groups. ASME Journal of Heat and Mass Transfer. 2020. Vol. 142. No. 6: 061806. Paper No. HT-19-1655. DOI: 10.1115/1.4046794

  10. Baumann M., Koch C., Staudacher S. Application of Neural Networks and Transfer Learning to Turbomachinery Heat Transfer. Aerospace, 2022, vol. 9, no. 2: 49. DOI: 10.3390/aerospace9020049

  11. Riegler C. Correlations to Include Heat Transfer in Gas Turbine Performance Calculations. Aerospace Science and Technology, 1999, vol. 3, no. 5, pp. 281–292. DOI: 10.1016/S1270-9638(00)86964-3

  12. Falaleev S.V., Badykov R.R., Matveev A.A., Grigor'ev E.M. Nasosy. Turbiny Sistemy, 2022, no. 4(45), pp. 46-56.

  13. Falaleev S.V.. Tisarev A.YU., Matveev A.A., Moshkov D.R. Materialy Mezhdunarodnoi nauchno-tekhnicheskoi konferentsii “Problemy i perspektivy razvitiya dvigatelestroeniYA” (23–25 June 2021; Samara). Samara, SSAU, 2021, pp. 66-67.

  14. Temis YU.M., Selivanov A.V., Yakushev D.A. Aviatsionnye dvigateli, 2021, no. 3(12), pp. 35-44. DOI: 10.54349/26586061_2021_3_35

  15. Gnizdor R.Y., Pyatykh I.N., Kaplin M.A., Rumyantsev A.V. Development and characteristics studying of the xenon and krypton operating SPD-70M thruster engineering model. Aerospace MAI Journal, 2023, vol. 30, no. 2, pp. 106-115. DOI: 10.34759/vst-2023-2-106-115

  16. Pyatunin K.R., Otryakhina K.V.. Kutsev N.M.. Lyaskin A.S. CADFEM Review, 2019, no. 6, pp. 77-81.

  17. Special Metals INCONEL Alloy 718, https://www.matweb.com/search/DataSheet.aspx?MatGUID=94950a2d209040a09b89952d45086134

  18. Riegler C. Modulares Leistungsberechnungsverfahren für Turboflugtriebwerke mit Kennfelddarstellung für Wärmeübertra-gungsvorgänge. PhD. Thesis, Universität Stuttgart, Düsseldorf, Germany, 1997.

  19. Simon V., Weigand B., Gomaa H. Dimensional Analysis for Engineers. Springer International Publishing, 2017, 224 p.

  20. JetCat PHT3 Einwellen Helikoptertriebwerk. URL: https://www.jetcat.de/de/productdetails/produkte/jetcat/produkte/hobby/pht3-4%20Getriebevarianten

mai.ru — informational site of MAI

Copyright © 1994-2024 by MAI