Method for estimating state of charge of battery through Kalman filtering

Abstract

A method for estimating the state of charge (SOC) of a battery through Kalman filtering is technologically characterized in that the SOC of a lithium battery pack is estimated by an extended Kalman filter (EKF) algorithm, a Vmin state space model of the lithium battery pack is built, and the minimum Vmin of the load voltage of the single batteries in the battery pack and the SOC of the battery pack as used as an observed variable and a state variable of the model. The adopted Kalman filter is an extended Kalman filter, SOC is worked out recursively by use of an ampere-hour integral method, SOC is substituted in an observation equation to get the estimated value of Vmin, the Kalman gain for every step is calculated, and the optimal estimate of SOC is worked out based on a state estimation observation renewal equation. The power battery SOC estimation method overcomes the defects of current error accumulation based on the ampere-hour integral method, and closed-loop estimation of state variable SOC is realized. As the effect of noise is taken into consideration in the process of calculation, the algorithm has a strong noise inhibition effect.

Description

technical field [0001] The invention relates to the technical field of electric vehicles, in particular to a method for estimating the state of charge (SOC, State of Charge) of a battery using Kalman filtering, which is applicable to all vehicles that need to use power batteries, especially power batteries that need to be estimated in real time SOC vehicle. [0002] Keywords: power battery electric vehicle SOC Background technique [0003] The electric vehicle battery pack exhibits a high degree of nonlinearity during use, and it is very difficult to accurately estimate the SOC. The SOC state of the battery pack is the basis of the power management system, and battery balancing and life state prediction require accurate SOC values. With the promotion of electric vehicles, how to accurately estimate the SOC value of the power battery pack has become a research hotspot in recent years. [0004] The currently commonly used SOC estimation method is to estimate the SOC by meas...

AI Technical Summary

Problems solved by technology

However, after a long learning time, a large amount of training data is required during this period, and the estimation error is greatly affected by the training data and learning methods.

Images