SOC estimation for Li-ion battery using optimum RLS method based on genetic algorithm

Abstract

Battery state of charge (SOC) has to be estimated properly in order to build a good battery management system (BMS) for electric vehicles. A Lithium battery dynamics has nonlinearity element and time varying parameter values. The speed of parameter change is different to each parameter. This paper proposes a new method of SOC estimation based on recursive least square (RLS) algorithm with multiple fixed forgetting factors whose optimum values are determined using Genetic Algorithm (GA). Open circuit voltage (OCV) is treated as parameter to be estimated together with internal resistance. Computer scripts of battery model, RLS, and GA have been built and numerical simulation has been conducted. Urban Dynamometer Driving Schedule (UDDS) was used as the input data. The results show that the combination algorithm between multiple fixed forgetting factor RLS (MFFF-RLS) and GA provides better performance than single fixed forgetting factor RLS (SFFF-RLS) as well as RLS without forgetting factor in estimating terminal voltage, OCV, SOC, and internal resistance of the Lithium battery. The multiple objective optimization can be solved using the built GA computer code giving performance index of 1.09*10 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">-6</sup> for SFFF-RLS and 2.11*10 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">-7</sup> for MFFF-RLS. It can be concluded that the proposed algorithm works satisfactory.