Closing the gap between finite difference and stirred reactor combustor modelling procedures

• Published in: Symposium, International, on Combustion Vol. 20; no. 1; pp. 541 - 547
• Main Authors: Ewan, B.C.R., Boysan, F., Swithenbank, J.
• Format: Journal Article
• Language: English
• Published: Elsevier Inc 1985
• ISSN: 0082-0784
• DOI: 10.1016/S0082-0784(85)80543-9

The objective of this study is to extend the scope of finite difference modelling of gas turbine combustors so that the maximum air loading (and also the rich and lean limits) can be predicted. Since the finite difference procedure is not suitable for the direct computation of blow off limits, it has been adapted to calculate the residence time distribution from which the equivalent stirred reactor of the primary stabilisation zone can be deduced. Well established stirred reactor modelling can then be used to compute the combustion stability loop and other features such as pollutant production and combustion efficiency. In this paper, a pulsed mercury tracer, which was detected optically, is used to measure the residence time distribution and this result is compared to that computed from our finite difference code. In the computation the turbulent trajectories of a large number of neutrally buoyant particles are used to represent the tracer pulse. Good agreement is found between the measured and the computed response, and the subsequently deduced maximum air loading of the combustor is consistent with the measured value for a variety of fuel nozzles.