Abstract:

Lithium-ion batteries are increasingly being used in fields such as satellites,portable devices and electric vehicles due to their high energy density and long service life. As an important index of the battery management system,accurate monitoring of the state of charge is important for ensuring the safety of battery use and improving battery efficiency. An adaptive center differential Kalman filtering algorithm is proposed for estimating the state of charge of the lithium-ion batteries. Firstly,this paper designs a linear Kalman filter to achieve the real-time estimations of the coefficients in the measurement equation,which avoids the testing of the relationship curve between the state of charge and the open circuit voltage. Secondly,considering that it is difficult to accurately obtain model parameters under certain operating conditions,the augmented vector method and adaptive central differential Kalman filter are used to achieve adaptive estimations of the state of charge and model parameters. Then,the linear Kalman filter and the adaptive center differential Kalman filter are coupled to achieve joint estimations of the state of charge,model parameters,and coefficients in the measurement equation,making the proposed algorithm better applicable to complex working conditions. In order to further improve the estimation accuracy and adaptability of the proposed algorithm to noises,the noise covariance matrices are dynamically adjusted via iterative method. Finally,the proposed algorithm is applied to different operating conditions to verify its effectiveness.

Key words: Lithium-ion battery, State of charge, Center differential Kalman filter, Adaptive estimation