An Ensemble of Recurrent Neural Networks for Real Time Performance Modelling of Three-Spool Aero-Derivative Gas Turbine Engine

Abstract Gas turbine is a complex system operating in non-stationary operation conditions for which traditional model-based modeling approaches have poor generalization capabilities. To address this, an investigation of a novel data-driven neural networks based model approach for a three-spool aero-derivative gas turbine engine (ADGTE) for power generation during its loading and unloading conditions is reported in this paper. For this purpose, a non-linear autoregressive network with exogenous inputs (NARX) is used to develop this model in MATLAB environment using operational closed-loop data collected from Siemens (SGT-A65) ADGTE. Inspired by the way biological neural networks process information and by their structure which changes depending on their function, multiple-input single-output (MISO) NARX models with different configurations were used to represent each of the ADGTE output parameters with the same input parameters. Usage of a single neural network to represent each of the system output parameters may not be able to provide an accurate prediction for unseen data and as a consequence, provides poor generalization. To overcome this problem, an ensemble of MISO NARX models is used to represent each output parameter. The major challenge of the ensemble generation is to decide how to combine results produced by the ensemble's components. In this paper, a novel hybrid dynamic weighting method (HDWM) is proposed. The simulation results show improvement in accuracy and robustness by using the proposed modeling approach.

Download Full-text

Development of Condition Monitoring Test Cell Using Micro Gas Turbine Engine

Volume 4: Cycle Innovations; Industrial and Cogeneration; Manufacturing Materials and Metallurgy; Marine ◽

10.1115/gt2009-59931 ◽

2009 ◽

Author(s):

Seonghee Kho ◽

Jayoung Ki ◽

Miyoung Park ◽

Changduk Kong ◽

Kyungjae Lee

Keyword(s):

Performance Analysis ◽

Real Time ◽

Gas Turbine ◽

Condition Monitoring ◽

Gas Turbine Engine ◽

Operating Conditions ◽

Test Cell ◽

Turbine Engine ◽

Time Performance ◽

Engine Condition

This study is aim to be programmed the simulation which is available for real-time performance analysis so that is to be developed gas turbine engine’s condition monitoring system with analyzing difference between performance analysis results and measuring data from test cell. In addition, test cell created by this study have been developed to use following applications: to use for learning principals and mechanism of gas turbine engine in school, and to use performance test and its further research for variable operating conditions in associated institutes. The maximum thrust of the micro turbojet engine is 137 N (14 kgf) at 126,000 rpm of rotor rotational speed if the Jet A1 kerosene fuel is used. The air flow rate is measured by the inflow air speed of duct, and the fuel flow is measured by a volumetric fuel flowmeter. Temperatures and pressures are measured at the atmosphere, the compressor inlet and outlet and the turbine outlet. The thrust stand was designed and manufactured to measure accurately the thrust by the load cell. All measuring sensors are connected to a DAQ (Data Acquisition) device, and the logging data are used as function parameters of the program, LabVIEW. The LabVIEW is used to develop the engine condition monitoring program. The proposed program can perform both the reference engine model performance analysis at an input condition and the real-time performance analysis with real-time variables. By comparing two analysis results the engine condition can be monitored. Both engine performance analysis data and monitoring results are displayed by the GUI (Graphic User Interface) platform.

Download Full-text