0
Research Papers

Model Validation for an Active Magnetic Bearing Based Compressor Surge Control Test Rig

[+] Author and Article Information
Se Young Yoon

Charles L. Brown Department of Electrical and Computer Engineering, University of Virginia, Charlottesville, VA 22904-4743

Zongli Lin1

Charles L. Brown Department of Electrical and Computer Engineering, University of Virginia, Charlottesville, VA 22904-4743

Kin Tien Lim, Christopher Goyne, Paul E. Allaire

Department of Mechanical and Aerospace Engineering, University of Virginia, Charlottesville, VA 22904-4746

1

Corresponding author.

J. Vib. Acoust 132(6), 061005 (Sep 16, 2010) (13 pages) doi:10.1115/1.4001845 History: Received November 18, 2009; Revised April 13, 2010; Published September 16, 2010; Online September 16, 2010

In this paper, we present experimental test data for the validation of a recently introduced mathematical model for centrifugal compression systems with variable impeller axial clearances. Employing the active magnetic bearings (AMBs) of a compressor built for the experimental study of surge, the axial clearance between the impeller and the static shroud is servo controlled, and the measured variations in the compressor output flow are compared with the mathematical model. The steady state and the dynamic responses of the compression system induced by varying the impeller tip clearance are measured and compared with the theoretical predictions, and the states of the compression system in surge condition are collected and analyzed. Parameters in the compression system model, such as the Greitzer parameter B and Helmholtz frequency ωH are experimentally identified. Also, the servo dynamics of the magnetic bearing that controls the axial impeller position is determined experimentally. To further validate the mathematical model and the feasibility of using the impeller tip clearance for controlling surge, we present a design example for an active surge controller based on the derived model, and simulate the response of the compression system. This design exercise also helps us understand the possible challenges that one could face in the design and implementation of a successful surge controller.

Copyright © 2010 by American Society of Mechanical Engineers
Your Session has timed out. Please sign back in to continue.

References

Figures

Grahic Jump Location
Figure 15

Block diagram of the weighted feedback system for the H∞ synthesis, where GC(s) is the compression system, W1(s) is the weight on plant output, W2(s) is the weight on the control input, W3(s) is the weight on the disturbance, W4(s) is the weight on the reference signal, and KC(s) is the active surge controller.

Grahic Jump Location
Figure 16

The throttle percentage opening, the compressor characteristic curve, the impeller tip clearance variation, and the plenum pressure rise variation for the augmented compression-AMB system with the surge controller in (21). The throttle is valve actuated from 37% to 28% opening.

Grahic Jump Location
Figure 17

The throttle percentage opening, the compressor characteristic curve, the impeller tip clearance variation, and the plenum pressure rise variation for the augmented compression-AMB system with the robust H∞ surge control. The throttle valve is actuated from 37% to 28% opening.

Grahic Jump Location
Figure 18

The throttle percentage opening, the compressor characteristic curve, the impeller tip clearance variation, and the plenum pressure rise variation for the augmented compression-AMB system with the robust H∞ surge control. The throttle valve opening with 1% sinusoidal disturbance.

Grahic Jump Location
Figure 19

The throttle percentage opening, the compressor characteristic curve, the impeller tip clearance variation, and the plenum pressure rise variation for the augmented compression-AMB system with the robust H∞ surge control.

Grahic Jump Location
Figure 1

Unshrouded single-stage centrifugal compressor with a vaneless diffuser, an electric motor, and a rotor supported on active magnetic bearings

Grahic Jump Location
Figure 2

Picture of the compressor, the inlet piping, and the modular exhaust piping

Grahic Jump Location
Figure 3

Drawing of the compressor surge control test rig layout, consisting of a single-stage unshrouded centrifugal compressor, a modular ducting system acting as a variable size plenum volume, and a throttle valve controlling the mass flow through the compression system

Grahic Jump Location
Figure 4

Experimentally determined compressor characteristic curve of exhaust pressure ratio versus mass flow rate, measured at different running speeds

Grahic Jump Location
Figure 5

Experimentally measured Bode plots for the open loop thrust AMB system, from the control voltage at the input of the AMB servo amplifiers to the rotor proximity sensor voltage output

Grahic Jump Location
Figure 6

Experimentally measured Bode plots for the closed-loop thrust AMB, from the rotor reference position voltage signal to the rotor proximity sensor voltage output. The thrust AMB system is stabilized with an H∞ feedback controller.

Grahic Jump Location
Figure 7

The tip clearance is the axial clearance between the impeller tip and the static shroud (35)

Grahic Jump Location
Figure 8

Polynomial fitting of the characteristic curve for the compression system operating at 16290 rpm

Grahic Jump Location
Figure 9

Experimentally obtained compressor characteristic curves at the nominal clearance and at the clearances of +6 mils and −6 mils. Also plotted for reference are the curves predicted from Eq. 3 at the nominal clearance (solid line), at +6 mils (dash-dotted line) and at −6 mils (dashed line), respectively.

Grahic Jump Location
Figure 10

Waterfall plot of experimental frequency response of the compressor pressure ratio to a perturbation in the impeller position of different frequencies. The impeller position was modulated at several input frequencies, and frequency components of the measured pressure ratio were plotted.

Grahic Jump Location
Figure 11

Experimental Bode plots from compressor tip clearance to plenum pressure rise at 35%, 32% and 30% throttle valve opening.

Grahic Jump Location
Figure 12

Experimental Bode plots from compressor tip clearance to plenum pressure rise versus simulation results of modified Greitzer’s compression system model (4). Throttle valve at 32% opening.

Grahic Jump Location
Figure 13

Plenum absolute pressure as the compressor is gradually driven from stable operation to surge.

Grahic Jump Location
Figure 14

Waterfall plot of the frequency components of the compressor pressure signal at different mass flow rates. The compressor enters surge at flows rates below 0.25 kg/s, with a dominating component at 21 Hz. As the mass flow rate is further reduced, a larger component at 7 Hz appears rapidly and dominates the frequency response.

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