An Overview of Methodologies to Predict Lean Blowout Limits for Gas Turbine Combustors

Abstract

The lean blowout (LBO) is a critical aspect of combustion performance for gas turbine combustors. During the past decades, three major prediction methodologies for the LBO limits, i.e. the semi-empirical model, the numerical prediction method and the hybrid prediction method are proposed. The semi-empirical models are derived mainly based on two kinds of physics-based models, i.e. the characteristic time (CT) model and the perfect stirred reactor (PSR) model. Among these semi-empirical models, Lefebvre’s LBO model that is based on the PSR model had been validated on 8 different aero gas turbine combustors with the prediction uncertainty ±30% and applied widely on the prediction of the LBO limits. Subsequently, a series of studies have been done to further develop Lefebvre’s LBO model. The numerical prediction methods are studied increasingly with the dramatically increase of the computing power. Based on the open literature, the best prediction uncertainty of the numerical prediction methods could be within 14% for a fixed combustor configuration with 3 kinds of fuels. More validations of different combustor configurations, atomization and dispersion models are required for the further application of numerical prediction methods. The hybrid prediction methods combine the semi-empirical models and the numerical methods simultaneously and could be divided into 2 types, i.e. the numerical and the semi-empirical based hybrid methods. The numerical based hybrid prediction method requires more validations and some general criteria for different configurations and operating conditions. The semi-empirical based hybrid prediction method could achieve maximum and average prediction uncertainties about ±15% and ±5%, respectively, for 10 combustor configurations. In summary, all the prediction methodologies should be further developed to achieve much more accurate prediction for the LBO limits as well as ensure the good generality.