Parameter identification of Li-Po batteries in electric vehicles: A comparative study

Abstract

An effective management of the onboard energy storage system is a key point for the development of electric vehicles. This requires the accurate estimation of the battery state over time and in a wide range of operating conditions. The battery state is usually expressed as state-of-charge and state-ofhealth. Its estimation demands an accurate model to represent the static and dynamic behaviour of the battery. Developing such a model requires the online identification of the battery parameters. This paper aims at comparing the performance of two popular system identification techniques, i.e., the Extended Kalman Filter and the classic Least Squares method. A significant contribution of this work is the definition of a benchmark which is representative of the real use of the battery in an electric vehicle. Simulation results show the peculiarities of both methods and their effectiveness.

Development of Accurate Lithium-Ion Battery Model Based on Adaptive Random Disturbance PSO Algorithm

“…The first type of methods is generally called traditional identification methods, like fitting method based on Least Squares [8], subspace identification [16], multiple linear regression method [18], and so on. This type of methods is simple and intuitive; however, the identified parameters have larger errors, and hence it is usually used in applications with lower accuracy demand [19,20]. The second type of methods is generally called bionic intelligent optimization algorithms, like PSO [21], genetic algorithm (GA), [22] and so on.…”

Open-circuit voltage measurement of Lithium-Iron-Phosphate batteries

“…Usually, two R-C branches with time constants in the range of seconds and minutes are sufficient to reproduce the battery dynamic behaviour in a way accurate enough for control purposes in a BMS [4], [5]. The values of the R-C parameters can be identified offline by means of electrochemical impedance spectroscopy [6] or pulsed current tests [4], [5] or during the battery operation using, for instance, the Extended Kalman filter or the Moving Window Least Squares method [7], [8].…”

Research on Parameter Identification of Battery Model Based on Adaptive Particle Swarm Optimization Algorithm

“…A comparative analysis of the advantages and disadvantages between MWLS and EKF is summarized in Ref. [13]. According to Ref.…”

“…According to Ref. [13], MWLS method is able to quickly trace the variations of parameter with defect of easy to get error under the nonlinear characteristic. However, EKF method is more stable with a slower tracing speed to variations of parameter.…”