0
Research Papers: Gas Turbines: Controls, Diagnostics, and Instrumentation

Multiple-Model Sensor and Components Fault Diagnosis in Gas Turbine Engines Using Autoassociative Neural Networks

[+] Author and Article Information
Z. N. Sadough Vanini

Department of Electrical
and Computer Engineering,
Concordia University,
Montreal, QC H3G 1M8, Canada

N. Meskin

Department of Electrical Engineering,
Qatar University,
Doha, Qatar
e-mail: nader.meskin@qu.edu.qa

K. Khorasani

Department of Electrical
and Computer Engineering,
Concordia University,
Montreal, QC H3G 1M8, Canada
e-mail: kash@ece.concordia.ca

Contributed by the Controls, Diagnostics and Instrumentation Committee of ASME for publication in the JOURNAL OF ENGINEERING FOR GAS TURBINES AND POWER. Manuscript received November 25, 2013; final manuscript received March 5, 2014; published online May 5, 2014. Assoc. Editor: Allan Volponi.

J. Eng. Gas Turbines Power 136(9), 091603 (May 05, 2014) (16 pages) Paper No: GTP-13-1429; doi: 10.1115/1.4027215 History: Received November 25, 2013; Revised March 05, 2014

In this paper the problem of fault diagnosis in an aircraft jet engine is investigated by using an intelligent-based methodology. The proposed fault detection and isolation (FDI) scheme is based on the multiple model approach and utilizes autoassociative neural networks (AANNs). This methodology consists of a bank of AANNs and provides a novel integrated solution to the problem of both sensor and component fault detection and isolation even though possibly both engine and sensor faults may occur concurrently. Moreover, the proposed algorithm can be used for sensor data validation and correction as the first step for health monitoring of jet engines. We have also presented a comparison between our proposed approach and another commonly used neural network scheme known as dynamic neural networks to demonstrate the advantages and capabilities of our approach. Various simulations are carried out to demonstrate the performance capabilities of our proposed fault detection and isolation scheme.

FIGURES IN THIS ARTICLE
<>
Copyright © 2014 by ASME
Your Session has timed out. Please sign back in to continue.

References

Zedda, M., and Singh, R., 2001, “Neural-Network-Based Sensor Validation for Gas Turbine Test Bed Analysis,” Proc. Inst. Mech. Eng. Part I: J. Syst. Control Eng, 215(1), pp. 47–56. [CrossRef]
Kiyak, E., Kahvecioglu, A., and Caliskan, F., 2011, “Aircraft Sensor and Actuator Fault Detection, Isolation, and Accommodation,” J. Aerosp. Eng., 24(1), pp. 46–58. [CrossRef]
Sarkar, S., Jin, X., and Ray, A., 2011, “Data-Driven Fault Detection in Aircraft Engines With Noisy Sensor Measurements,” ASME Eng. Gas Turbines Power, 133(8), p. 081602. [CrossRef]
Sheng, C. G. W., Napolitano, M., and Fravolini, M., 2002, “Online Learning RBF Neural Networks for Sensor Validation,” AIAA Guidance Navigation and Control Conference, Monterey, CA, August 5–8, AIAA Paper No. 2002-4996. [CrossRef]
Samy, I., Postlethwaite, I., and Gu, D., 2010, “A Comparative Study of NN and EKF Based SFDA Schemes With Application to a Nonlinear UAV Model,” Int. J. Control, 83(5), pp. 1025–1043. [CrossRef]
Neppach, C., and Casdorph, V., 1995, “Sensor Failure Detection, Identification and Accommodation in a System Without Sensor Redundancy,” 33rd Aerospace Sciences Meeting & Exhibit, January 1995, Reno, NV, AIAA Paper No. 95-0011.
Palm, T., Fast, M., and Thern, M., 2011, “Gas Turbine Sensor Validation Through Classification With Artificial Neural Networks,” Appl. Energy, 88(11), pp. 3898–3904. [CrossRef]
Joly, R., Ogaji, S., Singh, R., and Probert, S., 2004, “Gas-Turbine Diagnostics Using Artificial Neural-Networks for a High Bypass Ratio Military Turbofan Engine,” Appl. Energy, 78(4), pp. 397–418. [CrossRef]
Napolitano, M., and Windon, D., 1996, “A Comparison Between Kalman Filter and Neural Network Approaches for Sensor Validation,” AIAA Guidance, Navigation and Control Conference, San Diego, CA, July 29–31, AIAA Paper No. 96-3894. [CrossRef]
An, Y., 1998, “A Design of Fault Tolerant Flight Control Systems for Sensor and Actuator Failures Using On-Line Learning Neural Networks,” Ph.D. thesis, West Virginia University, Morgantown, WV.
Ogaji, S., Singh, R., and Probert, S., 2002, “Multiple-Sensor Fault-Diagnoses for a 2-Shaft Stationary Gas-Turbine,” Appl. Energy, 71(4), pp. 321–339. [CrossRef]
Ogaji, S., Sampath, S., Marinai, L., Singh, R., and Probert, S., 2005, “Evolution Strategy for Gas-Turbine Fault-Diagnoses,” Appl. Energy, 81(2), pp. 222–230. [CrossRef]
Marinai, L., Probert, D., and Singh, R., 2004, “Prospects for Aero Gas-Turbine Diagnostics: A Review,” Appl. Energy, 79(1), pp. 109–126. [CrossRef]
Fast, M., Assadi, M., and De, S., 2009, “Development and Multi-Utility of an ANN Model for an Industrial Gas Turbine,” Appl. Energy, 86(1), pp. 9–17. [CrossRef]
Rasaienia, A., Moshiri, B., and Moezzi, M., 2013, “Feature-Based Fault Detection of Industrial Gas Turbines Using Neural Networks,” Turkish J. Elec. Eng. Comput. Sci., 21, pp. 1340–1350. [CrossRef]
Ganguli, R., 2013, Gas Turbine Diagnostics, Signal Processing and Fault Isolation, CRC, Boca Raton, FL.
Gupta, S., Ray, A., Sarkar, S., and Yasar, M., 2008, “Fault Detection and Isolation in Aircraft Gas Turbine Engines. Part 1: Underlying Concept,” J. Aerosp. Eng., Part G, 222(3), pp. 307–318. [CrossRef]
Sarkar, S., Yasar, M., Gupta, S., Ray, A., and Mukherjee, K., 2008, “Fault Detection and Isolation in Aircraft Gas Turbine Engines. Part 2: Validation on a Simulation Test Bed,” J. Aerosp. Eng. Part G, 222(3), pp. 319–330. [CrossRef]
Palme, T., Fast, M., and Thern, M., 2011, “Gas Turbine Sensor Validation Through Classification With Artificial Neural Networks,” Thermal Power Eng., 88(11), pp. 3898–3904. [CrossRef]
Ayoubi, M., 1994, “Fault Diagnosis With Dynamic Neural Structure and Application to a Turbo Charger,” International Symposium on Fault Detection, Supervision and Safety for Technical Proceesses (SAFEPROCESS’94), Espoo, Finland, June 13–16, pp. 618–623.
Mohammadi, R., Naderi, E., Khorasani, K., and Hashtrudi-Zad, S., 2010, “Fault Diagnosis of Gas Turbine Engines by Using Dynamic Neural Networks,” ASME Paper No. GT2010-23586. [CrossRef]
Patan, K., and Parisini, T., 2005, “Identification of Neural Dynamic Models for Fault Detection and Isolation: The Case of a Real Sugar Evaporation Process,” J. Process Control, 15(1), pp. 67–79. [CrossRef]
Yazdizadeh, A., and Khorasani, K., 1997, “Identification of a Class of Nonlinear Systems Using Dynamic Neural Network Structures,” International Conference on Neural Networks, Houston, TX, June 9–12, pp. 194–198. [CrossRef]
Yazdizadeh, A., and Khorasani, K., 2002, “Adaptive Time Delay Neural Network Structures for Nonlinear System Identification,” Neurocomputing, 47(1–4), pp. 207–240. [CrossRef]
Tayarani-Bathaie, S. S., Sadough Vanini, Z. N., and Khorasani, K., 2013, “Dynamic Neural Network-Based Fault Diagnosis of Jet Engines,” Neurocompting, 125, pp. 153–165. [CrossRef]
Sadough Vanini, Z. N., Khorasani, K., and Meskin, N., 2013, “Fault Detection and Isolation of a Dual Spool Gas Turbine Engine Using Dynamic Neural Networks and Multiple Model Approach,” Inform. Sci., 259, pp. 234–251. [CrossRef]
Meskin, N., Naderi, E., and Khorasani, K., 2013, “A Multiple Model-Based Approach for Fault Diagnosis of Jet Engines,” IEEE Trans. Control Syst. Technol., 21(1), pp. 254–262. [CrossRef]
Naderi, E., Meskin, N., and Khorasani, K., 2012, “Nonlinear Fault Diagnosis of Jet Engines by Using a Multiple Model-Based Approach,” ASME J. Eng. Gas Turbines Power, 134(1), p. 011602. [CrossRef]
Naderi, E., Meskin, N., and Khorasani, K., 2010, “Fault Diagnosis of Jet Engine by Using a Multiple Model-Based Approach,” ASME Paper No. GT2010-23442. [CrossRef]
Hines, J., and Uhrig, R., 1998, “Use of Autoassociative Neural Networks for Signal Validation,” J. Intell. Robotic Syst., 21(2) pp. 143–154. [CrossRef]
Guo-Jian, H., Gui-Xiong, L., Geng-Xin, C., and Tie-Qun, C., 2010, “Self-Recovery Method Based on Auto-Associative Neural Network for Intelligent Sensors,” 8th World Congress on Intelligent Control and Automation (WCICA), Jinan, China, July 7–9, pp. 6918–6922. [CrossRef]
Najafi, M., Culp, C., and Langari, R., 2004, “Performance Study of Enhanced Auto-Associative Neural Networks for Sensor Fault Detection,” 4th International Conference for Enhanced Building Operations, Paris, October 18–19.
Guo, T., Mattern, D., Jaw, L., and Chen, C., 2003, “Model-Based Sensor Validation for a Turbofan Engine Using Autoassociative Neural Networks,” Int. J. Smart Eng. Syst. Design, 5(1), pp. 21–32. [CrossRef]
Kramer, M. A., 1998, “Autoassociative Neural Networks,” Comput. Chem. Eng., 16(4), pp. 313–328. [CrossRef]
Meskin, N., Naderi, E., and Khorasani, K., 2010, “Fault Diagnosis of Jet Engines by Using a Multiple Model-Based Approach,” ASME Paper No. GT2010-23442. [CrossRef]
Merrington, G. L., 1994, “Fault Diagnosis in Gas Turbines Using a Model-Based Technique,” ASME J. Eng. Gas Turbines Power, 116(2), pp. 374–380. [CrossRef]
Camporeale, S. M., Fortunato, B., and Mastrovito, M., 2006, “A Modular Code for Real Time Dynamic Simulation of Gas Turbines in SIMULINK,” ASME J. Eng. Gas Turbines Power, 128(3), pp. 506–517. [CrossRef]
Meskin, N., Naderi, E., and Khorasani, K., 2010, “Nonlinear Fault Diagnosis of Jet Engines by Using a Multiple Model-Based Approach,” ASME Paper No. GT2011-45143. [CrossRef]
Visser, W., Kogenhop, O., and Oostveen, M., 2004, “A Generic Approach for Gas Turbine Adaptive Modeling,” ASME J. Eng. Gas Turbines Power, 126(2), pp. 334–341. [CrossRef]
PROOSIS, 2014, “Ecosimpro and PROOSIS Simulation Tools,” EA Internacional, Madrid, Spain, http://www.ecosimpro.com/
Lu, P., and Hsu, T., 2002, “Application of Autoassociative Neural Network on Gas-Path Sensor Data Validation,” J. Propul. Power, 18(4), pp. 879–888. [CrossRef]

Figures

Grahic Jump Location
Fig. 1

Architecture of an autoassociative neural network (AANN) where σ denotes the sigmoidal nodes and l denotes the linear nodes

Grahic Jump Location
Fig. 2

The dual spool aircraft jet engine modules and information flow chart and interdependencies (for details refer to [21,28,35])

Grahic Jump Location
Fig. 3

Sensor fault detection scheme using the autoassociative neural network (AANN) architecture

Grahic Jump Location
Fig. 4

The parallel bank of AANN architecture proposed for performing the fault detection and isolation tasks simultaneously corresponding to L faulty modes and scenarios

Grahic Jump Location
Fig. 5

Input and output of the AANN0 network for noise filtering

Grahic Jump Location
Fig. 6

Error correction capability of the AANN0 scheme corresponding to sensor recovery and deviation rates for the TLC and PLC sensors

Grahic Jump Location
Fig. 7

The AANN1 sensor reconstruction and the residual signals for the sensor drift fault with the rate of 0.06% per second on the THT and N2 sensor measurements

Grahic Jump Location
Fig. 8

The AANN1 and AANN5 generated residuals corresponding to the case of a 4% ΔΓHT fault

Grahic Jump Location
Fig. 9

The AANN5, AANN6, and AANN8 generated residuals corresponding to the case of a 2% ΔηLT fault

Grahic Jump Location
Fig. 10

The AANN6 and AANN8 generated residuals corresponding to the case of a 3% ΔηLT fault

Grahic Jump Location
Fig. 11

The AANN2 and AANN5 generated residuals for the case of a 4% ΔηHT fault injected at t = 15 s and a 20 K bias injected on the THC sensor at t = 20 s

Grahic Jump Location
Fig. 12

Performance of the FDI scheme subject to concurrent faults. Top plots show the AANN7 residuals for the 5% ΔΓLT fault injected at t = 16 s and the 5% ΔηLT fault injected at t = 25 s. The bottom plots show the AANN8 residuals for the 5% ΔΓLT fault injected at t = 16 s and the 5% ΔηLT fault injected at t = 25 s.

Tables

Errata

Discussions

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related eBook Content
Topic Collections

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In