An active adaptive feedback control approach is utilized to minimize pressure fluctuations in a laboratory-scale dump combustor. Essential elements of the controller include sensors, actuators, and a control strategy. The actuation mechanism consists of spanwise forcing of the inlet boundary layer. This forcing is acoustically driven and has been shown to reduce the magnitude of the pressure oscillations by altering the dynamics of the flame-vortex interaction. The control strategy is based on the adaptive filtered-x least-mean-square (LMS) algorithm. Adaptive systems monitor their own performance and vary their internal structure to optimize this performance. The LMS algorithm does not require off-line gradient estimates to search for the minima on the performance surface, making it computationally efficient. A piezoelectric transducer measures the real-time pressure signal that is used as feedback as well as a measure of the error to train the adaptive algorithm. In its efforts to minimize the pressure fluctuations, the adaptive algorithm updates the weights of a digital filter, in real time. The digital filter, as well as the adaptive algorithm, resides on a digital signal processing (DSP) board. Both finite impulse response (FIR) and infinite impulse response (IIR) filter structures were investigated. It was found that the recursive direct-form filter structure is required to prevent the algorithm from driving itself toward instability.
[1]
Craig T. Bowman,et al.
Combustor performance enhancement through direct shear layer excitation
,
1990
.
[2]
William H. Press,et al.
Numerical recipes
,
1990
.
[3]
G. Billoud,et al.
Adaptive Active Control of Combustion Instabilities
,
1992
.
[4]
S. Candel,et al.
A review of active control of combustion instabilities
,
1993
.
[5]
Bernard Widrow,et al.
A comparison of adaptive algorithms based on the methods of steepest descent and random search
,
1976
.
[6]
J. D. Powell,et al.
An adaptive optimal combustion control strategy
,
1995
.
[7]
J. David Powell,et al.
On-line adaptive optimal combustor control
,
1996,
IEEE Trans. Control. Syst. Technol..
[8]
Bernard Widrow,et al.
Adaptive Signal Processing
,
1985
.