Uncertainty Quantification in the Conceptual Design of Hybrid-Electric General Aviation Aircraft

In the conceptual design stage, the uncertainties primarily stem from simplified modeling assumptions which can have a significant effect during the later stages of vehicle development. Typically, a simplified empirical and/or lower-fidelity aerodynamic model, traditional regression based weight estimation methods, and a rubberized engine-based propulsion system model are assumed during the initial sizing phase. Similarly, technology state-of-the-art is assumed to size propulsion system components such as batteries, generators, and motors. These variables are broadly classified as epistemic uncertainties. To demonstrate the uncertainty quantification process, a single-engine and a twin-engine general aviation aircraft are studied in this paper. Hybrid-electric propulsion architecture is proposed for the single-engine aircraft to enhance flight safety by allowing the propeller to be driven through two power paths: (1) a piston-engine power path and (2) a battery power path. For the twin-engine aircraft, an electrified propulsion architecture is presented that is capable of providing power to both propellers during one-engine inoperative flight conditions. Once the deterministic sizing results are compared with the baseline, this study investigates the impact of the uncertainties by varying the assumed model parameters and evaluating their influence on vehicle sizing results. A design-of-experiments (DOE) approach is employed to create a surrogate model, followed by a Monte-Carlo simulation to generate probability distributions of block fuel subject to aleatory variations and harmonic range subject to epistemic uncertainties. The simulation results are then used to compute the cumulative distribution function (CDF) and assess the sizing outcomes based on the CDF.