New method for estimating SoC of power lithium battery

Abstract

The invention discloses a new method for estimating a SoC of a power lithium battery. The method comprises the steps of: adopting an autoregressive exogenous (ARX) model to establish a lithium batteryequivalent model, then adopting a genetic algorithm based on an AIC criterion to determine order of the lithium battery equivalent model, and calculating model coefficients by utilizing a recursive least-squares method after determining the model order; establishing a spatial state equation required by Kalman filtering according to the acquired ARX model coefficients and a lithium battery state equation; and finally realizing estimation of the SoC of the lithium battery by utilizing a Kalman prediction iterative equation. The method just needs to determine the order and coefficients of the ARX model without calculating specific equivalent circuit parameter values, simplifies the tedious formula derivation and calculation process so as to avoid the subsequent calculation of e indexes of aKalman filtering state transfer matrix and an input control matrix, and reduces the error introduced because the equivalent circuit parameters are affected by complicated operating conditions of the lithium battery.

Description

technical field [0001] The invention relates to the field of estimating the state of charge of a battery, in particular to a new method for estimating the SoC of a power lithium battery. Background technique [0002] This year, the development of new energy electric vehicles has developed from the initial embryonic stage to the bottleneck stage, and the power management system (Battery Management System, BMS) has become a key technology that restricts its breakthrough development. The power management system mainly includes battery state of charge (SoC) estimation, battery temperature monitoring, and battery overvoltage, overcurrent, overcharge, overdischarge, and overtemperature protection. Among them, the battery SoC mainly completes the estimation of the remaining capacity of the battery and provides information such as cruising range for the driver. Due to the complex structure of the lithium battery itself, the work is affected by factors such as different working condi...

AI Technical Summary

Problems solved by technology

Due to the dynamic nonlinear characteristics of lithium batteries, and the influence of different working conditions, ambient temperature and other factors during the working process, it is difficult to reflect these factors into the required model parameters in the actual modeling process

Therefore, there are errors in the model parameters no matter whether offline identification or online identification is adopted, and the accumulation of errors will affect the accuracy of the established model and ultimately affect the estimation accuracy of SoC.

Images