IMPROVING DISTURBANCE COMPENSATION IN GAS TURBINES BY INCORPORATING EVENT-TRIGGERED LOGIC SIGNALS FROM SWITCHGEAR

摘要

Traditional feedback control schemes perform well when the gas turbine (GT) is operating near steady state operating points. In the case of large and abrupt changes in shaft power demand, for example, a standard feedback controller cannot offer the same performance as a better informed controller with more rapid detection time and lesser measurement delays. This work seeks to improve the quality of the information available for use in feedback control by incorporating event-triggered logic signals from switchgear. In this paper, we consider a GT generating power for a microgrid in island mode operation. We seek to use more immediate, but logic-valued, signals available from switchgear monitoring equipment, such as circuit breakers, to improve transient load demand estimation and thereby control performance. Techniques are developed using ideas familiar from Kalman filtering for rapid acquisition of load estimates during transients, notably faults. The resulting approach yields a time-varying estimator which uses the breaker logic signals to augment the standard feedback signals to improve transient estimator response without a concomitant degradation of steady state response. An example is provided using an identified closed-loop GT model.