Hardware-in-the-Loop Platform for Assessing Battery State Estimators in Electric Vehicles

Abstract

The development of new algorithms for the management and state estimation of lithium-ion batteries requires their verification and performance assessment using different approaches and tools. This paper aims at presenting an advanced hardware in the loop platform that uses an accurate model of the battery to test the functionalities of battery management systems (BMSs) in electric vehicles. The developed platform sends the simulated battery data directly to the BMS under test via a communication link, ensuring the safety of the tests. As a case study, the platform has been used to test two promising battery state estimators, the adaptive mix algorithm and the dual extended Kalman filter, implemented on a field-programmable gate array-based BMS. The results show the importance of the assessment of these algorithms under different load profiles and conditions of the battery, thus highlighting the capabilities of the proposed platform to simulate many different situations in which the estimators will work in the target application.