Methodology for short-term performance prognostic of gas turbine using recurrent neural network

The issue of performance prognosis has been a topic of considerable interest in industrial condition monitoring applications. An innovative data driven prognostic methodology has been introduced in the current study by utilizing artificial recurrent neural network (RNN) approach which intends to improve the capability of equipment performance prediction within a specified short time bound even with limited available data. The ability of the approach is demonstrated using condition monitoring parameters collected from a 20 MW industrial gas turbine. An appropriate selection and fusion of measured variables has been employed to feed RNN with the most influential performance information. The analysis demonstrated that the developed prognostic approach has a great potential to provide an accurate short term forecast of equipment performance which can be invaluable for maintenance strategy and planning.

[1]  K. Mathioudakis,et al.  Setting Up of a Probabilistic Neural Network for Sensor Fault Detection Including Operation With Component Faults , 2002 .

[2]  Masdi Muhammad,et al.  A Framework for Intelligent Condition-based Maintenance of Rotating Equipment using Mechanical Condition Monitoring , 2014 .

[3]  Carl S. Byington,et al.  An Overview of Selected Prognostic Technologies With Application to Engine Health Management , 2006 .

[4]  Yi-Guang Li,et al.  Gas turbine performance prognostic for condition-based maintenance , 2009 .

[5]  Tony Giampaolo,et al.  Gas Turbine Handbook: Principles and Practices , 2002 .

[6]  Les E. Atlas,et al.  Recurrent neural networks and robust time series prediction , 1994, IEEE Trans. Neural Networks.

[7]  Markus Wallin,et al.  A Comparative Study of Genetic Algorithms and Gradient Methods for RM12 Turbofan Engine Diagnostics and Performance Estimation , 2004 .

[8]  Joseph Mathew,et al.  Rotating machinery prognostics. State of the art, challenges and opportunities , 2009 .

[9]  Takashi Hiyama,et al.  Predicting remaining useful life of rotating machinery based artificial neural network , 2010, Comput. Math. Appl..

[10]  R. Bhargava,et al.  Condition Monitoring and Diagnostic Aspects of Gas Turbine Transient Response , 1992 .

[11]  Bin Zhang,et al.  Machine Condition Prediction Based on Adaptive Neuro–Fuzzy and High-Order Particle Filtering , 2011, IEEE Transactions on Industrial Electronics.

[12]  Meherwan P. Boyce,et al.  Gas turbine engineering handbook , 1981 .

[13]  Andrea Toffolo,et al.  Prediction of performance and emissions of a two-shaft gas turbine from experimental data , 2008 .

[14]  Daming Lin,et al.  A review on machinery diagnostics and prognostics implementing condition-based maintenance , 2006 .

[15]  Ranjan Ganguli,et al.  Genetic fuzzy system for damage detection in beams and helicopter rotor blades , 2003 .

[16]  Gabor Karsai,et al.  An evolvable tri-reasoner IVHM system , 2001, 2001 IEEE Aerospace Conference Proceedings (Cat. No.01TH8542).

[17]  G. Krishnaiah,et al.  Neural network approach for a combined performance and mechanical health monitoring of a gas turbine engine , 2012 .

[18]  Peng Wang,et al.  Fault prognostics using dynamic wavelet neural networks , 2001, Artificial Intelligence for Engineering Design, Analysis and Manufacturing.

[19]  Frank L. Lewis,et al.  Intelligent Fault Diagnosis and Prognosis for Engineering Systems , 2006 .

[20]  Lin Mi,et al.  Multi-steps degradation process prediction for bearing based on improved back propagation neural network , 2013 .

[21]  Masdi Muhammad,et al.  Power and Thermal Efficiency Study of Offshore Gas Turbines under Various Ambient Temperatures , 2015 .

[22]  Ming J. Zuo,et al.  Health condition prognostics of gears using a recurrent neural network approach , 2009, 2009 Annual Reliability and Maintainability Symposium.

[23]  Jeffrey L. Elman,et al.  Finding Structure in Time , 1990, Cogn. Sci..