State of Charge Estimation by Joint Approach With Model-Based and Data-Driven Algorithm for Lithium-Ion Battery
Qin Shi, Zhengxin Jiang, Zhi Wang, Xingguo Shao, Lin He
Abstract
In order to ensure the safety and service life of lithium-ion batteries for automotive applications, accurate state of charge is required for system management in the process of driving. It is particularly challenging to estimate the state of charge by using online approach, for which battery dynamics model is nonlinear. Many of researchers have focused on model-based or data-driven algorithms alone, but comparatively little of them use a joint approach with the two type of algorithms. The data-driven algorithm is self-learning and has better adaptability, while the model-based algorithm is more stable and has stronger robustness. If these advantages can be combined, a better state of charge estimation approach will be developed. In this paper, based on battery charge dynamics, complex fraction order model of battery is simplified into discrete fraction model for engineering application of control algorithm. A Bayesian belief network is used to estimate the battery model parameters, and the adaptive extended Kalman particle filter is used to estimate the state of charge. In order to obtain accurate parameters of battery model for training, linear programming is used to identify the parameters online. Collectively, this paper design a joint approach of how the adaptive extended Kalman particle filter with Bayesian belief network estimate the state of charge precisely. Developed approach has been downloaded into a battery control unit, and tested in real-world conditions using battery test bench to realize practical application of the joint approach.