Development of reduced order models for real-time helicopter flight simulation

Abstract

The scope and usefulness of a flight simulator goes well beyond its pilot training capabilities; indeed, simulators play an important role during the design process of vehicles and control systems. The ability of a simulator to accurately predict the behavior of a helicopter using as information only its physical characteristics would be highly desirable as it would allow manufacturers to get an early feedback from pilots on any design decision (concerning, for instance, handling qualities, rotorcraft-pilot coupling proneness, etc.). However, despite the complexity and the use of state-of-the-art components in modern simulators, they are not yet able to provide a fully coherent representation of reality. Moreover, with the aim of correcting some sub-optimal behavior in specific flight conditions and to respect the tolerances needed for the validation of a flight model, a certain amount of artificial tuning is often applied on top of the physical model. These modifications are often not justified from an engineering standpoint and, while improving simulations for particular operating conditions, they may have an adverse effect on other parts of the flight envelope. The need to tune the model can often be related to the deficiencies of the mathematical model describing the helicopter dynamics. The physics involved is indeed the result of the coupling of complex phenomena like the nonlinear structural dynamics of the slender main rotor blades, the complex rotor aerodynamic environment resulting from the combination of blade motion and inflow induced by wake vorticity remaining in close proximity of the rotor disk, the interaction of the air flow with the fuselage, the main and tail rotors and mutual interactions, the interaction with the ground, the dynamics of engine and actuators, the effects of control systems. Real-time simulation of these phenomena requires a suitable trade-off between modeling accuracy and computational efficiency. Modeling and simulation of the complete aerodynamic/aeroelastic response of a helicopter rotor during arbitrary manoeuvring flight conditions is yet far from being predicted with suitable accuracy. Research in the 1990's and 2000's in the USA pointed out the deficiencies in current rotor wake modeling for simulator applications and suggested that inaccurate and incomplete modeling of transient dynamics of the rotor wake results in deficiencies in simulator behavior to pilot control inputs. In addition, concerning rotorcraft pilot couplings (RPC), recent research highlighted the effects that aeroelastic and wake modeling may have on pilots biodynamic response. For these reasons the ability to include wake and aeroelastic effect in simulator models is fundamental. In this work, the focus is on the mathematical modeling of a main rotor aeroelastic operator suitable for simulators, and, more in general, on the development of a complete tool chain allowing to derive computationally effcient, reduced-order models, from complex aeroelastic solvers to be used for flight simulation tasks.