0
Research Papers

Self-Learning Based Centrifugal Compressor Surge Mapping With Computationally Efficient Adaptive Asymmetric Support Vector Machine

[+] Author and Article Information
Xin Wu

School of Energy, Power and Mechanical Engineering,  North China Electric Power University, No. 2 Beinong Road,Huilongguan, Beijing 102206, Chinawuxincn@gmail.com

Yaoyu Li1

Department of Mechanical Engineering,  University of Texas at Dallas, 800 W. Campbell Road, Richardson, TX 75080yaoyu.li@utdallas.edu

1

Corresponding author.

J. Dyn. Sys., Meas., Control 134(5), 051008 (Jul 27, 2012) (10 pages) doi:10.1115/1.4006219 History: Received November 26, 2009; Revised February 05, 2012; Published July 26, 2012; Online July 27, 2012

When an air compressor is operated at very low flow rate for a given discharge pressure, surge may occur, resulting in large oscillations in pressure and flow in the compressor. To prevent the damage of the compressor, on account of surge, the control strategy employed is typically to operate it below the surge line (a map of the conditions at which surge begins). Surge line is strongly affected by the ambient air conditions. Previous research has developed to derive data-driven surge maps based on an asymmetric support vector machine (ASVM). The ASVM penalizes the surge case with much greater cost to minimize the possibility of undetected surge. This paper concerns the development of adaptive ASVM based self-learning surge map modeling via the combination with signal processing techniques for surge detection. During the actual operation of a compressor after the ASVM based surge map is obtained with historic data, new surge points can be identified with the surge detection methods such as short-time Fourier transform or wavelet transform. The new surge point can be used to update the surge map. However, with increasing number of surge points, the complexity of support vector machine (SVM) would grow dramatically. In order to keep the surge map SVM at a relatively low dimension, an adaptive SVM modeling algorithm is developed to select the minimum set of necessary support vectors in a three-dimension feature space based on Gaussian curvature to guarantee a desirable classification between surge and nonsurge areas. The proposed method is validated by applying the surge test data obtained from a testbed compressor at a manufacturing plant.

Copyright © 2012 by American Society of Mechanical Engineers
Your Session has timed out. Please sign back in to continue.

References

Figures

Grahic Jump Location
Figure 3

Flow chart for self-learning surge map modeling process

Grahic Jump Location
Figure 4

Misclassification of data points: false alarm and unpredicted surge

Grahic Jump Location
Figure 5

ASVM based surge map modeling

Grahic Jump Location
Figure 6

Illustration for complexity reduction by reducing number of support vectors

Grahic Jump Location
Figure 7

The relation between convex or concave projection point and Gaussian curvature

Grahic Jump Location
Figure 8

Evolution of mild surge into deep surge for centrifugal air compressor

Grahic Jump Location
Figure 1

Surge phenomenon and surge line for dynamic compressors. (a) Pressure and flow oscillations during surge and (b) relationship between the surge limit line and the surge control line.

Grahic Jump Location
Figure 2

SVM model with high and low degree of convolution

Grahic Jump Location
Figure 9

PC/104 based surge detection for centrifugal air compressor

Grahic Jump Location
Figure 10

Stage-2 outlet pressure based surge detection via STFFT and CWT. (a) Stage-2 outlet pressure signal, (b) magnitude of 12 Hz component in STFFT of stage-2 pressure signal, and (c) temporal profile of 3rd-level Morlet son wavelet magnitude.

Grahic Jump Location
Figure 11

Surge evolution process during the operation of centrifugal air compressor. (a) Stage-2 outlet pressure signal, (b) STFFT results at 40 s, (c) STFFT results at 41 s, and (d) STFFT results at 42 s.

Grahic Jump Location
Figure 12

Plot of the actual surge and not-surge points for IGV opening and current

Grahic Jump Location
Figure 13

Distance of the selected support vectors to the hypersurface

Grahic Jump Location
Figure 14

Distance of validation data points to the hypersurface

Grahic Jump Location
Figure 15

Distance of all data points to the hypersurface

Grahic Jump Location
Figure 16

Distance of the selected support vectors to the hypersurface

Grahic Jump Location
Figure 17

Distance of validation data points to the hypersurface

Grahic Jump Location
Figure 18

Distance of all points to the hypersurface. (a) Pressure and flow oscillations during surge and (b) relationship between the surge limit line and the surge control line.

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