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

A Computationally Efficient Methodology for Generating Training Data for a Transient Neural Network of a Tip-Jet Reaction Drive System

[+] Author and Article Information
Brian K. Kestner1

 School of Aerospace Engineering, Georgia Institute of Technology, Atlanta, GA 30332-0150brian.kestner@asdl.gatech.edu

Jimmy C.M. Tai

 School of Aerospace Engineering, Georgia Institute of Technology, Atlanta, GA 30332-0150jimmy.tai@ae.gatech.edu

Dimitri N. Mavris

 School of Aerospace Engineering, Georgia Institute of Technology, Atlanta, GA 30332-0150dmavris@ae.gatech.edu

1

Corresponding author. Present address: School of Aerospace Engineering, Georgia Institute of Technology, Atlanta, GA 30332-0150.

J. Eng. Gas Turbines Power 133(12), 121601 (Aug 25, 2011) (11 pages) doi:10.1115/1.4003957 History: Received September 23, 2010; Revised March 22, 2011; Published August 25, 2011; Online August 25, 2011

This paper presents a computationally efficient methodology for generating training data for a transient neural network model of a tip-jet reaction drive system for potential use as an onboard model in a model based control application. This methodology significantly reduces the number of training points required to capture the transient performance of the system. The challenge in developing an onboard model for a tip-jet reaction drive system is that the model has to operate over the whole flight envelope, to account for the different dynamics present in the system, and to adjust to system degradation or potential faults. In addition, the onboard model must execute in less time than the update interval of the controller. To address these issues, a computationally efficient training methodology and neural network surrogate model have been developed that captures the transient performance of the tip-jet reaction system. As the number of inputs to a neural network becomes large, the computational time needed to generate the number of training points required to accurately represent the range of operating conditions of the system may become quite large also. A challenge for the tip-jet reaction drive system is to minimize the number of neural network training points, while maintaining the high accuracy. To address this issue, a novel training methodology is presented which first trains a steady-state neural network model and uses deviations from steady-state operating conditions to define the transient portion of the training data. The combined results from both the transient and the steady-state training data can then be used to create a single transient neural network of the system. The results in this paper demonstrate that a transient neural network using this new computationally efficient training methodology has the potential to be a feasible option for use as an onboard real-time model for model based control of a tip-jet reaction drive system.

Copyright © 2011 by American Society of Mechanical Engineers
Your Session has timed out. Please sign back in to continue.

References

Figures

Grahic Jump Location
Figure 1

Model based control architecture

Grahic Jump Location
Figure 2

Neural network schematic

Grahic Jump Location
Figure 3

Tip-jet reaction drive modes of operation [21]

Grahic Jump Location
Figure 4

“Cold” cycle tip-jet reaction drive system [22]

Grahic Jump Location
Figure 5

“Hot” or “warm” cycle tip-jet reaction drive system [22]

Grahic Jump Location
Figure 6

Tip-jet reaction drive system model

Grahic Jump Location
Figure 7

Notional steady-state and transient design space

Grahic Jump Location
Figure 8

Engine fuel flow step input

Grahic Jump Location
Figure 9

Response to a fuel flow step input

Grahic Jump Location
Figure 10

Engine fuel input at 0.6 rad/s

Grahic Jump Location
Figure 11

Response to a 0.6 rad/s engine fuel input

Grahic Jump Location
Figure 12

Engine fuel input at 6 rad/s

Grahic Jump Location
Figure 13

Response to a 6 rad/s engine fuel input

Grahic Jump Location
Figure 14

Engine fuel input at 60 rad/s

Grahic Jump Location
Figure 15

Response to a 60 rad/s engine fuel input

Grahic Jump Location
Figure 16

Multiple inputs with varying signals

Grahic Jump Location
Figure 17

Response to multiple inputs with varying signals

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