Investigation of Adaptive Signal Processing Approaches to Active Combustion Control.

摘要

There has been significant progress made in understanding dynamic models and control system designs for active control of thermoacoustic instabilities. During the past several years, there has been an increasing trend away from mostly empirical or experimentally-based active combustion control (ACC) methods in exchange for control systems that rely on more accurate understanding of the dynamic processes involved in the thermoacoustic limit cycling response. Early demonstrations of phase-shifting ACC designs (Gutmark et al., 1993; McManus et al., 1990) relied simply on a measurement of the acoustic pressure and actuation of the unsteady heat release after appropriate delay (i. e. phase shift) relative to the measured pressure signal. The phase-shifting controllers proved to be effective in many situations but were plagued by inadequacy knowledge of how to predict the required phase, and gain, of the controller for varying operating conditions of the combustor. These problems naturally led to investigations of adaptive control methods.