Abstract

Utilizing electrochemical energy storage (batteries) as a main unit of the propulsion system in electric aircraft is expected to be the dominant technology that offers more efficient and cleaner operation. Specifically, Lithium-Ion Batteries (LIB) capable of providing higher energy density and power at a lower cost are the most probable candidate to serve as a source of energy in the Electric Propulsion System (EPS). However, the safety, reliability, and durability of LIB is still controversial. Due to influences associated with design, production and adverse operational conditions, LIB safety might not be assured. To solve this deficiency and enhance safety, Battery Management System (BMS) is designed to monitor battery status and control its safe operation via measurement of the current, voltage and temperature at the cell, module and pack level. Smart BMS leverages a data-driven approach and artificial intelligence in which the battery’s internal dynamics is not directly considered can accurately estimate the battery states and implement robust control strategies. However, applying AI techniques such as machine learning and neural networks in BMS involves key challenges. The research aims to review the challenges of smart BMS and identify the research gaps to improve battery safety and reliability.

Keywords: Electric propulsion system; Lithium-ion battery; Battery reliability; Battery management system; State estimation; Artificial Intelligence

Abbreviations:AI: Artificial Intelligence; BMS: Battery Management System; EPS: Electric Propulsion System; LIB: Lithium-Ion Battery; PHM: Prognostics and Health Management; SoH: State of Health; TMS: Thermal Management System