Efficient Modeling of Hybrid-Electric Aircraft for Design and Performance Optimization Studies

Efficient Modeling of Hybrid-Electric Aircraft for Design and Performance Optimization Studies

Tran, Huy (TU Delft Aerospace Engineering)

Melkert, J.A. (mentor)

Delft University of Technology

Aerospace Engineering | Flight Performance and Propulsion

2020-12-15

The world is faced with two concurrent trends. The first is the growing demand for air travel in the long term. The second is the need to cut aircraft emissions due to environmental goals and ambitions set by authorities. Fully-electric aircraft are currently infeasible due to the limited battery specific energy and power. Hybrid-electric propulsion (HEP) technology offers an intermediate solution. HEP can work in tandem with new aircraft designs, which potentially leads to further reductions in fuel and energy consumption. This study aims to quantify those reductions. A parallel HEP aircraft mission simulation framework 'MASS' has been developed at Netherlands Aerospace Centre. Earlier studies with MASS have pointed out that HEP retrofit aircraft have the potential to reduce fuel consumption. The expectation is that even more fuel can be saved by varying the HEP parameters and airframe geometry together. Hence, this study focuses on extending MASS to incorporate new aircraft designs. The extension is made in Matlab and involves adding the capability to calculate aircraft level aerodynamics and weight based on the geometry. A tool called 'OpenVSP' is used to generate the aircraft geometry. Then, the aircraft drag polar is constructed using a combination of the aerodynamic solver 'VSPAero' and handbook methods. Maximum deviation between predicted and actual drag for the relevant lift coefficients is 7.3%. The weight is calculated using Class-II estimation methods. Two separate sets of formulae are used: One is meant for the tube-and-wing (TAW) configuration whereas the other is dedicated to blended wing-body (BWB). The predicted operational empty mass deviates less than 1% from the reference for TAW aircraft, and 2.3% for BWB. HEP-related research is currently focused on large passenger aircraft. In this study an A320neo aircraft is used as reference aircraft, which is parallel HEP retrofitted. It is subjected to hybrid-electric powertrain and geometry variations in order to quantify the potential fuel burn reductions. During the variation studies, the same mission of 800 nm and 150 pax is simulated. The selected powertrain design variables are the engine core scaling and the power split. The latter is the electric power provided by the batteries to the engine, divided by the total amount of power. The idea of using the power split is to support the downscaled engine core during high thrust phases, which would otherwise not have enough power. A design of experiments is conducted prior to the unconstrained optimization for minimal fuel burn. The optimization is performed directly using Matlab function 'fmincon'. The powertrain optimum shows a 3.1% fuel burn reduction. The geometry variation is done in two ways: one involves planform changes and the other a configuration change to BWB. The optimization with powertrain and planform variables is done via a surrogate model obtained using an artificial neural network. The optimum shows a 16.9% reduction in fuel burn, where the planform contributes more to the reduction than the powertrain. For the BWB fuel burn quantification the same process is applied. The powertrain optimization shows a 3% fuel burn reduction with respect to its non-HEP variant. A 13.8% fuel reduction is obtained with powertrain and planform variations. Again, the planform is a larger determinant of fuel burn reduction than the powertrain. Compared with the non-HEP A320neo, a maximum of 19.9% fuel burn reduction can be achieved.

hybrid electric propulsion
conceptual aircraft design
Modeling and optimisation
parallel hybrid electric
parameteric design

http://resolver.tudelft.nl/uuid:026e74d1-82dd-4b80-9640-97032e5341e4

Student theses

master thesis

© 2020 Huy Tran