0
Research Papers: Gas Turbines: Combustion, Fuels, and Emissions

Tuning of the Acoustic Boundary Conditions of Combustion Test Rigs With Active Control: Extension to Actuators With Nonlinear Response

[+] Author and Article Information
Mirko R. Bothien1

Institut für Strömungsmechanik und Technische Akustik, Technische Universität Berlin, 10623 Berlin, Germanymirko.bothien@tu-berlin.de

Christian Oliver Paschereit

Institut für Strömungsmechanik und Technische Akustik, Technische Universität Berlin, 10623 Berlin, Germany

1

Corresponding author.

J. Eng. Gas Turbines Power 132(9), 091503 (Jun 17, 2010) (10 pages) doi:10.1115/1.4000599 History: Received May 06, 2009; Revised October 21, 2009; Published June 17, 2010; Online June 17, 2010

In the design process, new burners are generally tested in combustion test rigs. With these experiments, as well as with computational fluid dynamics, finite element calculations, and low-order network models, the burner’s performance in the full-scale engine is sought to be predicted. Especially, information about the thermoacoustic behavior and the emissions is very important. As the thermoacoustics strongly depend on the acoustic boundary conditions of the system, it is obvious that test rig conditions should match or be close to those of the full-scale engine. This is, however, generally not the case. Hence, if the combustion process in the test rig is stable at certain operating conditions, it may show unfavorable dynamics at the same conditions in the engine. In previous works, the authors introduced an active control scheme, which is able to mimic almost arbitrary acoustic boundary conditions. Thus, the test rig properties can be tuned to correspond to those of the full-scale engine. The acoustic boundary conditions were manipulated using woofers. In the present study, an actuator with higher control authority is investigated, which could be used to apply the control scheme in industrial test rigs. The actuator modulates an air mass flow to generate an acoustic excitation. However, in contrast to the woofers, it exhibits a strong nonlinear response regarding amplitude and frequency. Thus, the control scheme is further developed to account for these nonlinear transfer characteristics. This modified control scheme is then applied to change the acoustic boundary conditions of an atmospheric swirl-stabilized combustion test rig. Excellent results were obtained in terms of changing the reflection coefficient to different levels. By manipulating its phase, different resonance frequencies could be imposed without any hardware changes. The nonlinear control approach is not restricted to the actuator used in this study and might therefore be of use for other actuators as well.

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 1

System with different acoustic boundary conditions induced by a change in geometry or by implementation of a liner

Grahic Jump Location
Figure 2

Schematic setup of the atmospheric test rig

Grahic Jump Location
Figure 3

Schematical setup of the control concept for manipulation of the acoustic boundary condition of a duct

Grahic Jump Location
Figure 4

Control schematic for impedance tuning at discrete frequencies

Grahic Jump Location
Figure 5

Network representation of the test rig

Grahic Jump Location
Figure 6

Magnitude of actuator transfer function versus excitation frequency and amplitude of control signal. Cold flow conditions with main air and mean actuator mass flow of 60 g/s, respectively. Each grid point corresponds to a measured value.

Grahic Jump Location
Figure 7

Phase of actuator transfer function versus excitation frequency and amplitude of control signal. The phase value for 0.5 V is subtracted for scaling. Cold flow conditions with main air and mean actuator mass flow of 60 g/s. Each grid point corresponds to a measured value.

Grahic Jump Location
Figure 8

Nonlinearity of actuator transfer function |GEPT| (scaled with |GEPT(0.5 V)|) versus amplitude of excitation signal. Measurements for two different excitation frequencies are shown: 102 Hz: ×; and 112 Hz: ○.

Grahic Jump Location
Figure 9

Look-up table for correction of actuator control signal. Depending on tuning frequency and calculated control signal, a corrected signal êset is written out.

Grahic Jump Location
Figure 10

Modified controller setup accounting for the nonlinear amplitude response characteristic of the actuator

Grahic Jump Location
Figure 11

Scheme to instantaneously detect the amplitude of a sinusoidal signal

Grahic Jump Location
Figure 12

Reflection coefficient for the baseline case without control (with orifice, solid ◻—without orifice, dashed ◇) and |Rcl(78 Hz)|=1 with φcl=−φus (dash-dotted ○). Top: magnitude; bottom: phase.

Grahic Jump Location
Figure 13

Spectra of acoustic pressure for the baseline case without control (with orifice, solid ◻—without orifice, dashed ◇) and |Rcl(78 Hz)|=1 with φcl=−φus (dash-dotted ○)

Grahic Jump Location
Figure 14

OH∗-chemiluminescence for the baseline case without control (with orifice, solid ◻—without orifice, dashed ◇) and |Rcl(78 Hz)|=1 with φcl=−φus (dash-dotted ○)

Grahic Jump Location
Figure 15

The rms value of OH∗-chemiluminescence of the flame (color code in counts). Left: uncontrolled with orifice; middle: uncontrolled without orifice; right: controlled with orifice—controller is adjusted to reproduce natural instability of the uncontrolled case without orifice.

Grahic Jump Location
Figure 16

Center of gravity lines of rms value of OH∗-chemiluminescence of the flame. Solid: uncontrolled with orifice; dashed: uncontrolled without orifice; dash-dotted: controlled with orifice—controller is adjusted to reproduce natural instability of the uncontrolled case without orifice.

Grahic Jump Location
Figure 17

Spectra of acoustic pressure for the baseline case without control (with orifice, solid ◻—without orifice, dashed ◇) and Rcl=1 exp(−iφus) for 72 Hz (dashed ▽) and 92 Hz (dash-dotted ×)

Grahic Jump Location
Figure 18

Spectra of OH∗-chemiluminescence for the baseline case without control (with orifice, solid ◻—without orifice, dashed ◇) and Rcl=1 exp(−iφus) for 72 Hz (dashed ▽) and 92 Hz (dash-dotted ×)

Grahic Jump Location
Figure 19

Reflection coefficient for the baseline case without control (with orifice, solid ◻—without orifice, dashed ◇) and Rcl=1 exp(−iφus) for 72 Hz (dashed ▽) and 92 Hz (dash-dotted ×). Asterisks mark the values to which the reflection coefficient’s phase should be tuned, i.e., −φus. Top: magnitude, bottom: phase.

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