0
TECHNICAL PAPERS: Gas Turbines: Controls, Diagnostics, and Instrumentation

A Stochastic Model for a Compressor Stability Measure

[+] Author and Article Information
Manuj Dhingra, Yedidia Neumeier, J. V. R. Prasad

 School of Aerospace Engineering, Georgia Institute of Technology, Atlanta, GA 30332

Andrew Breeze-Stringfellow, Hyoun-Woo Shin, Peter N. Szucs

 GE Aviation, Cincinnati, OH 45215

J. Eng. Gas Turbines Power 129(3), 730-737 (Nov 17, 2006) (8 pages) doi:10.1115/1.2718231 History: Received July 05, 2006; Revised November 17, 2006

A stability measure rooted in the unsteady characteristics of the flow field over the compressor rotor has been previously developed. The present work explores the relationship between the stochastic properties of this measure, called the correlation measure, and the compressor stability boundary. A stochastic model has been developed to gauge the impact of the correlation measure’s stochastic nature on its applicability to compressor stability management. The genesis of this model is in the fundamental properties of a specific stochastic process, one that is created by the threshold crossings of a random process. The model validation utilizes data obtained on three different axial compressor facilities. These include a single-stage low-speed axial compressor, a four-stage low-speed research compressor, and an advanced technology demonstrator high-speed compressor. This paper presents details of the model development and validation, as well as closed loop experimental results to demonstrate correlation measure’s usefulness in compressor stability management.

Copyright © 2007 by American Society of Mechanical Engineers
Your Session has timed out. Please sign back in to continue.

References

Figures

Grahic Jump Location
Figure 1

Expected performance enhancement with active control

Grahic Jump Location
Figure 2

GT-Axial results: The cumulative distribution of the correlation measure

Grahic Jump Location
Figure 3

LSRC results: The cumulative distribution of the correlation measure for compressor at 750rpm

Grahic Jump Location
Figure 4

LSRC results: The cumulative distribution of the correlation measure for compressor at 800rpm

Grahic Jump Location
Figure 5

Random occurrence of events. An event is defined as downwards crossing of a threshold.

Grahic Jump Location
Figure 6

GT-Axial results: The cumulative distribution of TBE. The threshold is set at the average correlation measure for each case.

Grahic Jump Location
Figure 17

Importance of threshold parameter Cth for surge limit avoidance. A poor choice coupled with stochastic nature of alarms can lead to failure of the controller.

Grahic Jump Location
Figure 16

GT-Axial rig results. The open-pause-close controller is able to avoid surge under throttle transients.

Grahic Jump Location
Figure 15

GT-Axial rig results. The simple open-pause-close control law keeps the system close to its maximum pressure operation.

Grahic Jump Location
Figure 14

HSC results: The cumulative distribution of TBE for the third stage. The threshold Cth is set 0.56.

Grahic Jump Location
Figure 13

HSC results: The cumulative distribution of TBE for the second stage. The threshold Cth is set 0.62.

Grahic Jump Location
Figure 12

HSC results: The cumulative distribution of TBE for first stage. The threshold Cth is set 0.82.

Grahic Jump Location
Figure 11

LSRC results: The cumulative distribution of TBE for compressor at 800rpm. The threshold Cth is set 0.92.

Grahic Jump Location
Figure 10

LSRC results: The cumulative distribution of TBE for compressor at 750rpm. The threshold Cth is set 0.86.

Grahic Jump Location
Figure 9

GT-Axial results: The day to day variation of the observed average number of events. The labels next to each marker designate the error in stall margin as estimated via least-square error curve fit.

Grahic Jump Location
Figure 8

Variation of the average number of events with stall margin. The threshold value plays a dominant role in this relationship.

Grahic Jump Location
Figure 7

GT-Axial results: The cumulative distribution of TBE with Cth of 0.76. The experimentally observed distribution (markers) is compared to that predicted by the proposed model (solid lines).

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