Estimation of Lithium Battery State of Charge by Fusion Algorithm of Forgetting Factor Multi-innovation Least Squares and Extended Kalman Filter

作者：Zhang, Fujian; Ye, Weidong^*; Lei, Guoping; Liu, Yingying; Wang, Xian

来源：SENSORS AND MATERIALS, 2022, 34(4): 1471-1485.

DOI：10.18494/SAM3659

• lithium battery
• state of charge
• forgetting factor multi-innovation least squares
• extended Kalman filter
• heterogeneous integration

Lithium-ion power batteries have been widely used in electric vehicles, micro electromechanical systems (MEMS), and integrated circuits owing to their high energy and long life. The state of charge (SoC) of the power batteries is an important parameter of electric vehicles, which directly affects the safety control and range of electric vehicles. In this paper, we propose a method of accurately estimating the lithium battery SoC in the presence of sensor bias and varying environmental temperature by the fusion algorithm based on the forgetting factor multi-innovation least squares (FF-MILS) and extended Kalman filter (EKF). To establish an accurate ternary lithium battery model and monitor the SoC of the battery online, a second order RC equivalent circuit model was used to fit the relationship between the open-circuit voltage (OCV) and the SoC through experimental data. The online FF-MILS algorithm was used to identify the model parameters of lithium batteries, which were introduced into the fading factor EKF algorithm. To alleviate the decline in accuracy caused by the sensor deviation, the sensor deviation was set in the fading factor EKF state-space model by using a sensor deviation cooperative estimation method. In a MATLAB/Simulink simulation, the proposed fusion algorithm quickly converged to the initial error, and the maximum error was less than 2% in the stable state, thus verifying the estimation accuracy of the proposed fusion algorithm and its robustness to sensor deviation and changes in ambient temperature.