Modelling of Lithium-ion battery and SOC estimation using simple and extended discrete Kalman Filters for Aircraft energy management

Abstract

As an energy storage unit Lithium-ion batteries play an important role in More Electrical Aircraft and accurate knowledge of its behaviour and state of charge (SOC) are required to ensure the safety of the Aircraft Electrical System (AES) and its energy management. This paper details the conception of a 2nd order equivalent electrical circuit (EEC) battery model with SOC, current (magnitude and direction) and temperature dependant parameters. A 50Ah LiFePO4 battery is subjected to Hybrid Pulse Power Characterisation (HPPC) for parameter identification and constant current discharge tests for model validation. The model shows good performance between 100% & 10% SOC with a maximum error of less than ±1% Subsequently, for SOC estimation, the Discrete Simple Kalman Filter (KF) is applied with the Open-Circuit Voltage (OCV) depending linearly on SOC, I and T. Secondly the Discrete Extended KF (EKF) is applied with the model parameters and OCV varying non-linearly with SOC. Their performance is then compared to that of the Coulomb-counting method with the EKF showing good tracking performance: converging to the true SOC given false initial values and the mean SOC estimation error reducing from ±7% with KF to less than ±1% with EKF.