Improving the energy storage capacity system in electric aircraft as well as overall load capacity is instrumental in urban air mobility. The EU-funded OPENSRUM project focuses on improving the power conversion efficiency in electric aircraft and designing a smart wireless battery management system with accurate battery state-of-charge (SoC) and state-of-health (SoH) estimations. The project will increase the voltage of the lithium-ion battery packs to reduce current rating and cable weight, while identifying a power converter topology to maximise overall efficiency. OPENSRUM will explore data-driven methods based on machine learning to improve the accuracy of the SoC and SoH estimations and reduce the gap between peak error and root mean square error.
The research proposal addresses the design challenges in the power conversion and the energy storage systems in the electric aircraft used for urban air mobility (UAM). The success of UAM as an alternate transportation system is strongly dependent on designing the overall system to be safe, efficient and reliable. This proposal focuses on improving the power conversion efficiency and designing a smart wireless battery management system (BMS) with accurate battery state-of-charge (SoC) and state-of-health (SoH) estimations. Another desirable aspect in the UAM aircraft is improving the overall payload capacity, which is impacted by the weight of the batteries, interconnection wiring and power conversion efficiency. The proposal aims to improve it by increasing the voltage of the Li-ion battery packs above the current state-of-the-art, which would reduce the current rating and cable weight, while identifying a power converter topology to maximize the overall efficiency. The design optimisation will consider the impacts of higher insulation requirement with higher voltages and overall cost. The power converter topology and the accompanying filters are optimised to reduce electromagnetic interference that can affect the sensitive electronics on the aircraft. The proposal explores data-driven machine-learning based methods to improve the accuracy of the SoC and SoH estimations and reduce the gap between peak error and the root-mean-square error (RMSE). A reduction in the gap between peak and RMSE will provide a reliable upper bound unlike for the case when estimation methods show a lower RMSE but a wide variation in the peak error. The wireless BMS will provide the advantage of easier maintenance and elimination of the conventional wiring weight.
This is a timely and innovative project that will help in novel technology development for UAM industry. It will help the applicant gain additional technical and managerial skills that would ensure a successful research career.