Approaches to Increase the Accuracy of Classification of the Compressor Blading Erosion Degree

Abstract

Erosive wear of the axial compressor blading reduces efficiency of the gas turbine units and could lead to their decommissioning. One of the methods in assessing and predicting the erosive wear under the gas turbine operation conditions is introduction of the machine learning models. The use of compressor operation parameters as the indicators makes it possible to determine the probable blading defect degree. For example, it was previously established that accuracy in classifying the axial compressor blading erosion degree by the machine learning methods and using initial results of the numerical experiments with the NASA Stage 37 transonic model stage reached 80 %. In continuing this stage research, the paper provides approaches to improve accuracy in solving the problem of classifying the erosive wear degree of the compressor blading. The paper identifies four approaches to improve the erosion prediction accuracy using the machine learning methods. They include reducing classification categories by expanding the wear degree values range for each category, adding artificial data in learning by interpolating results of the numerical experiments, solving the erosion classification problem by the regression methods, and creating an ensemble of the best models. Combination of the proposed methods results in increasing accuracy in the blading wear prediction up to 87--97 %

The work was supported by the RSF (project no. 22-79-00169)

Please cite this article in English as:

Blinov V.L., Deryabin G.A., Zubkov I.S. Approaches to increase the accuracy of classification of the compressor blading erosion degree. Herald of the Bauman Moscow State Technical University, Series Mechanical Engineering, 2024, no. 4 (151), pp. 129--142 (in Russ.). EDN: YTDSCQ

References

[1] Kurz R., Brun K. Degradation of gas turbine performance in natural gas service. J. Nat. Gas Sc. Eng., 2009, vol. 1, no. 3, pp. 95--102. DOI: https://doi.org/10.1016/j.jngse.2009.03.007

[2] Krivosheev I.A., Kamaeva R.F., Strugovets S.A. Features traffic dust particles in the flowing part and changes in geometry compressor blades in the operation of gas turbines. Vestnik UGATU, 2011, no. 3, pp. 18--24 (in Russ.). EDN: PWTNPZ

[3] De Pratti G.M. Aerodynamical performance decay due to fouling and erosion in axial compressor for GT aeroengines. E3S Web Conf., 2020, vol. 197, art. 11002. DOI: https://doi.org/10.1051/e3sconf/202019711002

[4] Tabakoff W., Lakshminarasimha A.N., Pasin M. Simulation of compressor performance deterioration due to erosion. J. Turbomach., 1990, vol. 112, no. 1, pp. 78--83. DOI: https://doi.org/10.1115/1.2927424

[5] Cumpsty N.A. Compressor aerodynamics. Malabar, Krieger, 2004.

[6] Gumerov A.V., Akmaletdinov R.G. Compressor blade erosion modeling. Vestnik SGAU im. akademika S.P. Koroleva [Vestnik of Samara State Aerospace University named after academician S.P. Korolev], 2011, vol. 2, no. 3-2, pp. 233--239 (in Russ.). EDN: OWQPXT

[7] Dvirnik Ya.V., Pavlenko D.V. Evaluation of compressor blades limit state of helicopter gas turbine engines depended on natural vibration frequencies. Aviatsionno-kosmicheskaya tekhnika i tekhnologiya [Aerospace Technic and Technology], 2016, no. 7, pp. 79--83 (in Russ.).

[8] Blinov V.L., Zubkov I.S. Assessment of the erosion wear influence on the operational parameters of the gas turbine unit’s axial compressor with CFD. Polet [Flight], 2023, no. 1-2, pp. 51--60 (in Russ.). EDN: CAKVBS

[9] Blinov V.L., Zubkov I.S., Bogdanets S.V., et al. Studies of erosive wear of the blading in axial compressors of gas turbines. Teploenergetika, 2023, no. 6, pp. 41--55 (in Russ.).DOI: https://doi.org/10.56304/S0040363623060024

[10] Cavarzere A., Venturini M. Application of forecasting methodologies to predict gas turbine behavior over time. J. Eng. Gas Turbines Power, 2012, vol. 134, no. 1, art. 012401. DOI: https://doi.org/10.1115/1.4004184

[11] Rajendran R. Gas turbine coatings --- an overview. Eng. Fail. Anal., 2012, vol. 26, pp. 355--369. DOI: https://doi.org/10.1016/j.engfailanal.2012.07.007

[12] Liu Y., Ravichandran R., Chen K., et. al. Application of machine learning to solid particle erosion of APS-TBC and EB-PVD TBC at elevated temperatures. Coatings, 2021, vol. 11, no. 7, art. 845. DOI: https://doi.org/10.3390/coatings11070845

[13] Zahedi P., Parvandeh S., Asgharpour A., et al. Random forest regression prediction of solid particle erosion in elbows. Powder Technol., 2018, vol. 338, pp. 983--992. DOI: https://doi.org/10.1016/j.powtec.2018.07.055

[14] Zarate J.E., Lopez M.A.G., Troyo J.A.C. Analysis and detection of erosion in wind turbine blades. Math. Comput. Appl., 2022, vol. 27, no. 1, art. 5. DOI: https://doi.org/10.3390/mca27010005

[15] Cao Q., Chen S., Zheng Y., et al. Classification and prediction of gas turbine gas path degradation based on deep neural networks. Int. J. Energy Res., 2021, vol. 45, no. 7, pp. 10513--10526. DOI: https://doi.org/10.1002/er.6539

[16] Blinov V.L., Deryabin G.A., Zubkov I.S. Machine learning introduction to classify the erosive wear degree of the compressor stage blades. Herald of the Bauman Moscow State Technical University, Series Mechanical Engineering, 2023, no. 4 (147), pp. 88--105 (in Russ.). EDN: MDWUNM

[17] Joshuva A., Sugumaran V. A machine learning approach for condition monitoring of wind turbine blade using autoregressive moving average (ARMA) features through vibration signals: a comparative study. Prog. Ind. Ecol., 2018, vol. 12, no. 1-2, pp. 14--34. DOI: https://doi.org/10.1504/PIE.2018.095867

[18] Reid L., Moore R.D. Design and overall performance of four highly loaded, high speed inlet stages for an advanced high-pressure-ratio core compressor. Report NASA-TP-1337. Cleveland, NASA Lewis Research Center, 1978.

[19] Blinov V.L., Zubkov I.S. Verification of a transonic stage CFD model for assessing the erosion wear influence on the operation of the axial compressor. Vestnik Samarskogo universiteta. Aerokosmicheskaya tekhnika, tekhnologii i mashinostroenie [Vestnik of Samara University. Aerospace and Mechanical Engineering], 2023, vol. 22, no. 1, pp. 51--62 (in Russ.). DOI: https://doi.org/10.18287/2541-7533-2023-22-1-51-62

[20] Pedregosa F., Varoquaux G., Gramfort A., et al. Scikit-learn: machine learning in Python. J. Mach. Learn. Res., 2011, vol. 12, pp. 2825--2830. DOI: https://doi.org/10.48550/arXiv.1201.0490