基于气液动力学模型的多算法融合SOC估算

Multi-algorithm fusion SOC estimation based on gas-liquid dynamics model

基于气液动力学模型,提出了一种多算法融合SOC估算方法,以提高电池在不同动态工况下的SOC估算精度与可靠性。相比传统卡尔曼滤波(Kalman filter,KF)算法,无迹卡尔曼滤波(unscented Kalman filter,UKF)算法通过对称采样与非线性点变换的方法,有着先进性和时变性,该工作将UKF观测器应用于气液动力学电池模型并进行SOC估算,并与神经网络(back propagation neural network,BPNN)相结合对估算结果做出修正,从而进一步提高估算精度。最后在不同工况下对方法进行了验证。结果表明,与原始算法相比,多算法融合SOC估算方法的精度和可靠性得到了提高。

A multi algorithm fusion SOC estimation method based on gas-liquid dynamics model was proposed to improve the accuracy and reliability of SOC estimation for batteries under different dynamic conditions.Compared with the traditional Kalman filter(KF)algorithm,the unscented Kalman filter(UKF)algorithm was progressiveness and time-varying through the method of symmetric sampling and nonlinear point transformation.In this work,the UKF observer was applied to the gasliquid dynamic battery model for SOC estimation,and further combining with neural networks(BPNN)to make corrections to the results,thereby further improving the estimation accuracy.Finally,the method was validated under different operating conditions.The results show that compared with the original algorithm,the accuracy and reliability of the multi algorithm fusion SOC estimation method have been improved.