Skip to main content
SpringerLink
Log in

Dynamic neural networks for gas turbine engine degradation prediction, health monitoring and prognosis

  • Predictive Analytics Using Machine Learning
  • Published:
Neural Computing and Applications Aims and scope Submit manuscript

Abstract

In this paper, the problem of health monitoring and prognosis of aircraft gas turbine engines is considered by using computationally intelligent methodologies. Two different dynamic neural networks, namely the nonlinear autoregressive with exogenous input neural networks and the Elman neural networks, are developed and designed for this purpose. The proposed dynamic neural networks are designed to capture the dynamics of two main degradations in the gas turbine engine, namely the compressor fouling and the turbine erosion. The health status and condition of the engine in terms of the turbine output temperature (TT) are then predicted subject to occurrence of these deteriorations. Various scenarios consisting of fouling and erosion separately as well as combined are considered. For each scenario, several neural networks are trained and their performance in predicting multiple flights ahead TTs is evaluated. Finally, the most suitable neural networks for achieving the best prediction are selected by using the normalized Bayesian information criterion model selection. Simulation results presented demonstrate and illustrate the effective performance of our proposed neural network-based prediction and prognosis strategies.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8
Fig. 9
Fig. 10
Fig. 11
Fig. 12
Fig. 13
Fig. 14
Fig. 15
Fig. 16

Similar content being viewed by others

References

  1. Naeem M, Singh R, Probert D (1998) Implications of engine deterioration for fuel usage. Appl Energy 29(1):125–146

    Article  Google Scholar 

  2. Brotherton T, Jahns G, Jacobs J, Wroblewski D (2000) “Prognosis of faults in gas turbine engines”

  3. Heng A, Zhang S, Tan AC, Mathew J (2009) Rotating machinery prognostics: state of the art, challenges, and opportunities. Mech Syst Signal Process 23(3):724–739

    Article  Google Scholar 

  4. Pusey HC (2007) Turbomachinery condition monitoring and failure prognosis. SAVIAC/Hi-Test Laboratories, Winchester, Virginia

    Google Scholar 

  5. Visser WPJ, Kogenhop O, Oostveen M (2006) A generic approach for gas turbine adaptive modeling. J Eng Gas Turbines Power 128(1):13–19

    Article  Google Scholar 

  6. Li YG (2002) Performance analysis based gas turbine diagnostics: a review. J Power Energy 216(5):363–377

    Article  Google Scholar 

  7. Naderi E, Meskin N, Khorasani K (2011) Nonlinear fault diagnosis of jet engines by using a multiple model based approach. J Eng Gas Turbines Power 134(1), Art. ID 011602. doi:10.1115/1.4004152

  8. Aker GF, Saravanamuttoo HI (1989) Predicting gas turbine performance degradation due to compressor fouling using computer simulation techniques. J Eng Gas Turbines Power 111(2):343–400

    Article  Google Scholar 

  9. Visser WP, Broomhead MJ (2000) Gsp: a generic object-oriented gas turbine simulation environment. National Aerospace Laboratory (NLR)

  10. Kobayashi T, Simon DL (2003) Application of a bank of Kalman filters for aircraft engine fault diagnostics. ASME Turbo Expo 461–470. doi:10.1115/GT2003-38550

  11. Meskin N, Naderi E, Khorasani K (2010) Fault diagnosis of jet engines by using a multiple model-based approach. ASME Turbo Expo 319–329. doi:10.1115/GT2010-23442

  12. Zedda M, Singh R (1998) Fault diagnosis of a turbofan engine using neural-networks: a quantitative approach. 34th AIAA, ASME, SAE, ASEE Joint Propulsion Conference & Exhibit, AIAA, pp 98–3602

  13. Kanelopoulos ASK, Mathioudakis K (1997) Incorporating neural-networks into gas-turbine performance diagnostics. ASME 97-GT-35, International Gas-turbine & Aero-engine Congress & Exhibition

  14. Ogaji S (2003) Advanced gas-path fault diagnostics for stationary gas-turbines. PhD Thesis, School of Engineering, Cranfield University

  15. Green A (1997) Artificial-intelligence for real-time diagnostics and prognostics of gas-turbine engines. AIAA 97-2899 Conference

  16. Vachtsevanos G, Wang P (2001) Fault prognosis using dynamic wavelet neural network

  17. Parker BE, Nigro TM, Carley MP, Barron RL, Ward DG, Poor HV, Rock D, DuBois TA (1993) Helicopter gearbox diagnostics and prognostics using vibration signature analysis, 1965. In: Proceedings of SPIE 1965, Applications of artificial neural networks IV, 2 Sept 1993, pp 531–542. doi:10.1117/12.152553

  18. Huang R, Xi L, Li X, Liu CR, Qiu H, Lee J (1998) Implications of engine deterioration for fuel usage. Appl Energy 59(2):125–146

    Google Scholar 

  19. Lee J (2007) A systematic approach for developing and deploying advanced prognostics technologies and tools: methodology and applications. Harrogate. http://www.sematech.org/meetings/archives/mfg/8492/pres/J_Lee.pdf

  20. Jianzhong S, Hongfu Z, Haibin Y, Pecht M (2010) Study of ensemble learning-based fusion prognostics. In: PHM prognostics and health management conference, PHM ’10, IEEE, Macao, 12–14 Jan 2010, pp 1–7

  21. Xue G, Xiao L, Bie M, Lu S (2005) Fault prediction of boilers with fuzzy mathematics and RBF neural network. Commun Circuits Syst 2:1012–1016

    Google Scholar 

  22. Gebraeel NZ, Lawley MA (2008) A neural network degradation model for computing and updating residual life distributions. IEEE Trans Autom Sci Eng 5(1):154–163

    Article  Google Scholar 

  23. Dragomir O, Gouriveau R, Zerhouni N (2008) Adaptive neuro-fuzzy inference system for mid term prognosis error stabilization. Int J Comput Commun Control 1:6

    Google Scholar 

  24. Lee J, Ni J, Djurdjanovic D, Qiu H, Liao H (2006) Intelligent prognostics tools and e-maintenance. Comput Ind 54(1):476–489

    Article  Google Scholar 

  25. Wang WQ, Golnaraghi MF, Ismail F (2004) Prognosis of machine health condition using neuro-fuzzy systems. Mech Syst Signal Process 18(4):813–831

    Article  Google Scholar 

  26. Zhao F, Tian Z, Zeng Y (2013) Uncertainty quantification in gear remaining useful life prediction through an integrated prognostics method. IEEE Trans Reliab 62(1):146–159

    Article  Google Scholar 

  27. Ishikawa M, Moriyama T (1996) Prediction of time series by a structural learning of neural networks. Fuzzy Sets Syst 82(2):167–176

    Article  Google Scholar 

  28. Torella G, Lombardo G (1995) Utilization of neural-networks for gas-turbine engines. XII ISABE 95–703

  29. Volponi RGA, DePold H, Daguang C (2000) The use of Kalman-filter and neural-network methodologies in gas-turbine performance diagnostics: a comparative study. ASME 2000-GT-547

  30. Han M, Fan J, Wang J (2011) A dynamic feedforward neural network based on gaussian particle swarm optimization and its application for predictive control. IEEE Trans Neural Netw 22(9):1457–1468

    Article  Google Scholar 

  31. Eddahech A, Briat O, Bertrand N, Deltage J-Y, Vinassa J-M (2012) Behavior and state-of-health monitoring of li-ion batteries using impedance spectroscopy and recurrent neural networks. Int J Electr Power Energy Syst 42(1):487494

    Article  Google Scholar 

  32. Cao Q, Ewing BT, Thompson MA (2012) Forecasting wind speed with recurrent neural networks. Eur J Oper Res 221(1):148154

    Article  MathSciNet  MATH  Google Scholar 

  33. Shen Z, Zhao K (2014) Dynamic neural network identification and decoupling control approach for MIMO time-varying nonlinear systems. Abstract and Applied Analysis, vol. Article ID 316206, 10 pages

  34. Mrugalski M (2014) Designing of dynamic neural networks. Advanced Neural Network-Based Computational Schemes for Robust Fault Diagnosis, Studies in Computational Intelligence 510

  35. Yazdizadeh A, Khorasani K (2002) Adaptive time delay neural network structures for nonlinear system identification. Neurocomputing 47:207–240

    Article  MATH  Google Scholar 

  36. Ayoubi M (1994) Fault diagnosis with dynamic neural structure and application to a turbocharger. Proc. Int. Symp. Fault Detection, Supervision and Safety for technical Process SAFEPROCESS’94

  37. Al-Zyoud IA-D, Khorasani K (2005) Detection of actuator faults using a dynamic neural network for the attitude control subsystem of a satellite. Proceedings of International Joint Conference on Neural Networks

  38. Valdes KKA, Ma L (2009) Dynamic neural network-based fault detection and isolation for thrusters in formation flying of satellites. Adv Neural Netw - ISNN 2009: 6th International Symposium on Neural Networks

  39. Patan K, Parisini T (2005) Identification of neural dynamic models for fault detection and isolation: the case of a real sugar evaporation process. IFAC J Process Control 15(1):67–79

    Article  Google Scholar 

  40. Mohammadi KKR, Naderi E, Hashtrudi-Zad S (2010) Fault diagnosis of gas turbine engines by using dynamic neural networks. Proc ASME Turbo Expo

  41. Jardine A, Lin D, Banjevic D (2006) A review on machinery diagnostics and prognostics implementing condition-based maintenance. Mech Syst Signal Process 20(7):1483–1510

    Article  Google Scholar 

  42. Contreras LR, Modi C, Pennathur A (2002) Integrating simulation modelling and equipment condition diagnostics for predictive maintenance strategies—a case study. Proc Winter Simul Conf

  43. Escher PC (1995) Pythia: an object-oriented gas path analysis computer program for general applications. Cranfield University, Cranfield

    Google Scholar 

  44. Luo J, Pattipati K, Qiao L, Chigusa S (2008) Model-based prognostic techniques applied to a suspension system. IEEE Trans Syst Man Cybern 38(5):1156–1168

    Article  Google Scholar 

  45. Endrenyi J, Aboresheid S, Allan RN (2001) The present status of maintenance strategies and the impact of maintenance on reliability. IEEE Trans Power Syst 16(4):638–646

    Article  Google Scholar 

  46. Asmai SA, Hussin B, Yusof MM (2010) A framework of an intelligent maintenance prognosis tool. Second IEEE International Conference on Computer Research and Development, pp 241–245

  47. Narendra KS, Parthasarathy K (1990) Identification and control of dynamical systems using neural networks. IEEE Trans Neural Netw 1(1):4–27

    Article  Google Scholar 

  48. Qin SZ, Su HT, McAvoy TJ (1992) Comparison of four neural net learning methods for dynamic system identification. IEEE Trans Neural Netw 3(1):122–130

    Article  Google Scholar 

  49. Su HT, McAvoy TJ, Werbos P (1992) Long-term predictions of chemical processes using recurrent neural networks: a parallel training approach. Ind Eng Chem Res 31(5):1338–1352. doi:10.1021/ie00005a014

    Article  Google Scholar 

  50. Chen S, Billings SA, Grant PM (1990) Non-linear system identification using neural networks. Int J Control 51(6):1191–1214

    Article  MathSciNet  MATH  Google Scholar 

  51. Amani P, Robertsson A (2011) Narx-based multi-step ahead response time prediction for database servers. In: Proceedings of 11th international conference on intelligent systems design and applications (ISDA), pp 813–818

  52. Diaconescu E (2008) The use of narx neural network to predict chaotic time series. WSEAS Trans Comput Res 3(3):182–191

    Google Scholar 

  53. Shen HY, Chang LC (2012) On-line multistep-ahead inundation depth forecasts by recurrent narx networks. Hydrol Earth Syst Sci Dis 9(10):11999–12028

    Article  Google Scholar 

  54. Cai L, Ma SY, Zhou YL (2010) Prediction of SYM-H index during large storms by NARX neural network from IMF and solar wind data. Ann Geophys 28:381–393

    Article  Google Scholar 

  55. Thissen U, Van Brakel R, de Weijer AP, Melssen WJ, Buydens LMC (2003) Using support vector machines for time series prediction. Chemometr Intell Lab Syst 69(1):35–49

    Article  Google Scholar 

  56. Liu C, Jiang D, Zhao M (2009) Application of RBF and Elman neural networks on condition prediction in CBM. In: Proceedings of the Sixth International Symposium on Neural Networks. Springer, pp 847–855

  57. Yang B, Mu X, Zhang Q (2010) Elman neural network based temperature prediction in cement rotary kiln calcining process. IEEE

  58. Sjoberg J, Zhang Q, Ljung L, Benveniste A, Delyon B, Glorennec P, Hjalmarsson H, Juditsky A (1995) Non-linear black-box modelling in system identification: a unified overview. Automatica 31(12):1691–1724

    Article  MathSciNet  MATH  Google Scholar 

  59. Horne B, Gilles C (1995) An experimental comparison of recurrent neural networks. Neural Inf Process Syst 7:697–704

    Google Scholar 

  60. Gao Y, Er M (2005) Narmax time series model prediction: feedforward and recurrent fuzzy neural network approaches. Fuzzy Sets Syst 150(2):331–350

    Article  MathSciNet  MATH  Google Scholar 

  61. Jiang C, Song F (2011) Sunspot forecasting by using chaotic time-series analysis and narx network. J Comput 6(7):1424–1429

    Google Scholar 

  62. Elman JL (1990) Finding structure in time. Cogn Sci 14(2):179–211

    Article  Google Scholar 

  63. Zhang J, Chung H, Lo W (2008) Chaotic time series prediction using a neuro-fuzzy system with time-delay coordinates. IEEE Trans Knowl Data Eng 20(7):956–964

    Article  Google Scholar 

  64. Zhang J, Chung H, Lo W (2002) Combining neural network model with seasonal time series ARIMA model. Technol Forecast Soc Change 69(1):71–87

    Article  Google Scholar 

  65. Zhang GP (2003) Time series forecasting using a hybrid arima and neural network model. Neurocomputing 50(6):159–175

    Article  MATH  Google Scholar 

  66. Lin CJ, Chen CH (2005) Identification and prediction using recurrent compensatory neuro-fuzzy systems. Fuzzy Sets Syst 150(2):307–330

    Article  MathSciNet  MATH  Google Scholar 

  67. Xie H, Tang H, Liao Y (2009) Time series prediction based on narx neural networks: An advanced approach. Proceedings of 8th International Conference on Machine Learning and Cybernetics

  68. Pham HT, Tran VT, Yang BS (2010) A hybrid of non-linear autoregressive model with exogenous input and autoregressive moving average model for long-term machine state forecasting. Expert Syst Appl 37(4):3310–3317

    Article  Google Scholar 

  69. Armstrong JS (2001) Identification of asymmetric prediction intervals through causal forces. J Forecast 20(4):273–283

    Article  Google Scholar 

  70. Meng F, Meeker WQ (2011) Coverage properties of weibull prediction interval procedures to contain a future number of failures. Iowa State University, Ames, IA

    Google Scholar 

  71. (2012) e-handbook of statistical methods nist/sematech. http://www.itl.nist.gov/div898/handbook/April

  72. Ogaji S, Sampath S, Singh R, Probert SD (2002) Parameter selection for diagnosing a gas-turbines performance-deterioration. Appl Energy 73(1):25–46

    Article  Google Scholar 

  73. Urban LA (1975) Parameter selection for multiple fault diagnostics of gas turbine engines. J Eng Power 97(2):225–230. doi:10.1115/1.3445969

  74. Camporeale SM, Fortunato B, Mastrovito M (2006) A modular code for real time dynamic simulation of gas turbines in simulink. J Eng Gas Turbines Power 128(3):506–517

    Article  Google Scholar 

  75. Kurz R, Brun K (2000) Degradation in gas turbine systems. Proceedings of the International Gas Turbine & Aeroengine Congress & Exhibition, Munich, Germany

  76. Flesland SM (December 2010) Gas turbine optimum operation. Master of Science in Product Design and Manufacturing, Norwegian University of Science and Technology

  77. Zwebek A (2002) Combined cycle performance deterioration analysis. PhD Thesis, Cranfield University

  78. Meher-Homji CB, Bromley A (2004) Gas turbine axial compressor fouling and washing. Proceedings of the Turbomachinery Symposium

  79. Acquah H (2010) Comparison of Akaike information criterion (aic) and Bayesian information criterion (bic) in selection of an asymmetric price relationship. Dev Agric Econ 2(1):1–6

    Google Scholar 

  80. Burnham KP, Anderson DR (2004) Multimodel inference understanding aic and bic in model selection. Sociol Methods Res 33(2):261–304

    Article  MathSciNet  Google Scholar 

  81. Posada D, Buckley TR (2004) Model selection and model averaging in phylogenetics: advantages of akaike information criterion and bayesian approaches over likelihood ratio tests. Syst Biol 53(5):793–808

    Article  Google Scholar 

  82. Hirose K, Kawano S, Konishi S, Ichikawa M (2011) Bayesian information criterion and selection of the number of factors in factor analysis models. J Data Sci 9(2):243–259

    MathSciNet  Google Scholar 

  83. Schwarz E, Gideon E (1978) Estimating the dimension of a model. Ann Stat 6(2):461–464

    Article  MathSciNet  MATH  Google Scholar 

Download references

Author information

Authors and Affiliations

  1. Department of Electrical and Computer Engineering, Concordia University, Montreal, QC, Canada

    S. Kiakojoori & K. Khorasani

Authors
  1. S. Kiakojoori
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. K. Khorasani
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to K. Khorasani.

Additional information

This publication was made possible by NPRP Grant No. 4-195-2-065 from the Qatar National Research Fund (a member of Qatar Foundation). The statements made herein are solely the responsibility of the authors.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Kiakojoori, S., Khorasani, K. Dynamic neural networks for gas turbine engine degradation prediction, health monitoring and prognosis. Neural Comput & Applic 27, 2157–2192 (2016). https://doi.org/10.1007/s00521-015-1990-0

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s00521-015-1990-0

Keywords

Search

Navigation