FIELD: engines and pumps.
SUBSTANCE: invention relates to a combustion system and a method for predicting the concentration of pollutants of a combustion system for a gas turbine engine. Aim of the invention is to provide a more reliable predictive emission control system. Combustion system (100) comprises combustion chamber (101) into which pilot fuel (102) and main fuel (103) is injected and ignited, wherein exhaust gas (104) generated by burned pilot fuel (102) and burned main fuel (103) is exhausted out of combustion chamber (101). Control unit (112) is coupled to fuel control unit (105) for adjusting pilot fuel (102) ratio, with temperature sensor (107) for receiving a temperature signal, with fuel determination unit (109) for receiving a specific fuel signal, and with mass flow sensor (110) for receiving a specific mass flow signal. Control unit (112) is adapted for determining a predicted pollutant concentration of exhaust gas (104) on the basis of a temperature signal, a fuel signal, a mass flow signal and a fuel split ratio. Method for predicting the concentration of pollutants of a combustion system for a gas turbine engine comprises several stages. Fuel control unit (105) divides the fuel (106) into pilot fuel (102) and main fuel (103). Fuel is injected and ignited inside the combustion chamber. Temperature signal is generated by temperature sensor (107) indicating the temperature of the exhaust gas inside combustion chamber (101) or downstream of the combustion chamber. Mass flow signal is generated by mass flow sensor (110) indicating mass flow rate (111) of the air entering the combustion chamber. Fuel determination unit (109) determines a fuel signal indicative of the fuel composition (106). Control unit based on the temperature signal, the fuel signal, the mass flow rate signal, and the fuel separation ratio determine the predicted concentration of exhaust gas pollutants (104).
EFFECT: prediction of the concentration of pollutants of a combustion system for a gas turbine engine.
11 cl, 3 dwg
Authors
Dates
2018-01-23—Published
2014-10-27—Filed