PHYSICS BASED MODELING OF A HELICOPTER GAS-TURBINE PROPULSION
SYSTEM FOR TRAINING SIMULATION
Ding-Jen “Dean” Liu
Simulation and Research Services Division, SAIC
Hampton, Virginia
Troy D. Abbott
Simulation and Research Services Division, SAIC
Lexington Park, Maryland
ABSTRACT
Mathematical modeling of complex, non-linear physical systems has often been limited by computational
requirements. Simple tools have resulted in the development and use of simplified models, making use of gross
approximations and focus on “effects modeling”, where the final output of the model is more important than
modeling of the physical properties which govern the behavior of the system. This approach has been observed
on many training simulations, which were among the first devices requiring real-time computation, and hence
required simplification of complex systems in order to function with limited computational capability. Today,
however, though computational horsepower continues to grow exponentially, many trainer systems continue to
rely on simplified “effects based” modeling approaches. Limitations of these models are not always obvious,
since they can often be tuned to match a specific set of criteria data very well. Without a foundation in physics,
however, these models quickly lose fidelity when operated “off condition”, or when malfunctions are
introduced.
The development of a physically representative model of the aircraft’s engines has resulted in increased
trainer fidelity, while reducing the amount of software required. Whereas previous model required “special”
operation for startup, shutdown, and malfunctions, the improved engine model is able to replicate engine
performance throughout the operational envelop using the same set of core engine component models during
the entire run cycle. Further, because each mechanical component of the engine and its control system are
modeled in a physically representative manner, malfunctions may be inserted at the component level, allowing
appropriate failure responses to cascade through the system.
This paper first presents the shortfalls of the previous propulsion system model. A very brief introduction to
gas-turbine engine thermal dynamics is followed by examination of the UH-1N engine and engine control
components and the methods by which they are modeled. Methods of resolving unknown engine parameters
(compressor-turbine matching) are then presented, followed by verification and comparison of the results with
the actual engine performance.
