Steps towards mixed energy optimisation of Uncrewed Aerial Vehicles

Abstract

The long-range uncrewed aerial vehicle (UAV) is expected to utilise a combination of power sources, including photovoltaic modules, fuel cells, or direct combustion with hydrogen, lithium compound batteries, and hydrocarbon fuels. Hybrid powertrains are becoming more common, highlighting the need for automated powertrain topological exploration to identify the optimal design for mission performance. In this thesis, we develop tools to identify such optimal designs. The first building block of such an optimiser is a scheme capable of the flexible, futureproof, yet economical encoding of powertrain topologies, and we propose such a scheme in this thesis. This scheme is adaptable for analysis and can be applied to any UAV powertrain topology, providing a means to compare and analyse different powertrains. The second building block of the optimisation process is determining the optimal energy architecture. To achieve this, we use a point mass model to represent the aircraft performance, the energy components are modelled and the different powertrain topologies are evaluated through dynamically generated strings, based on the proposed encoding scheme. The point mass model can be used to evaluate different powertrain topologies for fixed-wing UAVs through the initial value problem (IVP) by simulating the dynamics of the UAV’s motion over time. By simulating the UAV’s motion over time, the model can provide insights into the energy consumption and other performance metrics of different powertrain topologies. We compare two different fixed wing UAV models: a 5 kg maximum take-off mass with a 2mwing span and a 25 kg maximum take-off mass with a 4mwing span UAV; for set missions with different powertrains. Overall, we demonstrate that the framework proposed here delivers the lightest feasible powertrain topologies, describing the power supply as well as the preliminary flight profile for a specific platform for a given operation whilst developing a scheme that is suitable for embedding various topology strings with capability for a wide variety of multi-source and multi sink power systems. The proposed framework allows for refining the mission profile, minimising powertrain sizing, and finding an optimal energy split for hybrid powertrain topologies through novel powertrain topology parameterisation.

More information

Identifiers

ORCID for Andras Sobester: orcid.org/0000-0002-8997-4375

Catalogue record

Export record