On-line Lithium-Ion Battery State-of-Power Prediction by Twice Recursive Method Based on Dynamic Model Tongji University

Format:

Book

Conference/Event

Author/Creator:

Wang, Wang, author.

Contributor:

Dai, Haifeng

Wei, Xuezhe

Conference Name:

WCX SAE World Congress Experience (2019-04-09 : Detroit, Michigan, United States)

Language:

English

Physical Description:

1 online resource cm

Place of Publication:

Warrendale, PA SAE International 2019

Summary:

AbstractState-of-Power (SoP) prediction of Li-ion battery is necessary in battery management system for electric vehicles in order to deal with limited conditions, prevent overcharge and over discharge situations, increase the life of the battery and provide effective battery operation. This article suggests a method to on-line predict the 10-s charge and discharge peak power of Li-ion battery by twice recursions. First with the dynamic battery model we use the first recursion based on a least square method to get parameters which are influenced by the state of charge of Li-ion battery and temperature, et cetera The dynamic model is an equivalent circuit model. Current and voltage are input online into the battery model. By recursive least square method the parameters are updated in real time. Moreover, when we use a recursive method to get real-time parameters, we add an extra proper factor to abandon old datum, which increases the real-time capability of state-of-power prediction. By assuming a constant current input and using the dynamic model we get the present dynamic voltage. Then by the second recursion, we derive the formula of 10-s resistance and calculate the SoP which can last for 10 seconds. The variables of the formula are the parameters which we get directly from the first recursion. Without using the parameters to calculate ohmic resistance, polarization resistance or capacitance of battery, it reduces much calculation amount and improves the calculation speed. This method is validated with datum from NEDC tests of Li-ion battery. The 10-s resistance values are predicted accurately. The method is suitable for the application in the battery management system of electric vehicles

Notes:

Vendor supplied data

Publisher Number:

2019-01-1311

Access Restriction:

Restricted for use by site license