Modeling of Start-Up From Engine-Off Conditions Using High Fidelity Turbofan Engine Simulations

Engine models are widely used to simulate the engine behavior at steady state and transient operating conditions over the full flight envelope. Within the engine development process such simulations are used to support component design, evaluate engine performance, operability and test data, as well as to develop and optimize the engine controls. Recent developments have raised interest in the modeling of start-up processes of turbofan engines in order to support the definition of sufficient engine control laws. This implies that simulations are started at a condition where the engine shafts are static and temperatures and pressures are equal to ambient. During start-up the engine can only be operated transiently through the sub-sub-idle region (near zero speed) using a starter torque. The activity presented here was targeted to support the development of industrial-standard high-fidelity turbofan engine models capable of simulating start-up, shutdown or windmilling operation. Within the three previously mentioned cases starting from an engine-off condition, ground start from zero-speed is the most challenging in terms of physical and numerical modeling. For this reason, this paper concentrates on that case only. Zero mass flow and speed at the beginning of the simulation impose a set of special problems that do not exist in standard simulations: the modeling of a static engine-off condition, the modeling of static friction, and the modeling of reverse flows. The requirement to support an existing industrial model development process also made it necessary to apply the same quality of physical modeling to start-up simulations as would be the case for above-idle engine simulations. The physical effects present during engine start are discussed and modeling solutions are presented. Finally, results of a dry crank simulation are presented and discussed, illustrating that the expected effects are present and that the simulation is capable of predicting the correct trends.