Research on multilayer fast equalization strategy of Li-ion battery based on adaptive neural fuzzy inference system

Abstract

In this paper, an adaptive neuro fuzzy inference system (ANFIS) based multilayer fast equalization strategy for Li-ion batteries is developed to solve the problem of wrong fuzzy rule settings affecting the equalization speed in traditional fuzzy logic methods. Firstly, a multilayer equalization circuit is designed to perform both intra-group and inter-group equalization, and a Graph theory analysis shows that this circuit topology can perform fast equalization while ensuring better equalization efficiency. Then an adaptive neuro fuzzy inference system is designed based on the multilayer equalization circuit designed in this paper to achieve more accurate adaptive adjustment of the equalization current and reduce time consumption, and finally a comparative study of the equalization circuit and equalization method under static and charging/discharging conditions is conducted using the Matlab/Similink platform. The experimental study shows that the speed of the multilayer equalization circuit designed in this study is increased by about 62 % and 37.55 % compared with the traditional single-inductor equalization circuit and adjacent-inductor circuit, respectively. Compared with the conventional fuzzy logic control (FLC) equalization strategy, the speed of the proposed equalization strategy is improved by about 10.98 %. It effectively speeds up the equalization process of Li-ion batteries.