Method for estimating battery by strong tracking cubature Kalman filtering based on noise interference

Abstract

The invention provides a method for estimating a battery through strong tracking cubature Kalman filtering based on noise interference, based on noise interference of the battery and strong tracking Cubature Kalman Filter (CKF), estimate of the remaining electric quantity of the battery is performed, and the method includes the following characteristics: 1) analyzing the law of charge and discharge of battery charges, and obtaining an estimation theory of the SOC (State Of Charge) of the battery; 2) establishing a battery SOC model considering internal influencing factors such as battery temperature and a charge and discharge rate, and adopting a first-order RC equivalent model to establish a battery state equation; 3) performing online estimation of the SOC of the battery by a strong tracking cubature Kalman filtering algorithm; and 4) judging whether noise interference exists, if yes, performing self-adaptive square root cubature Kalman filtering to filter out noise, and if not, directly estimating the SOC of the battery. The method for estimating a battery through strong tracking cubature Kalman filtering based on noise interference has a relatively good estimation effect of estimating the remaining electric quantity of a battery model, and overcomes the defect that an error of a battery circuit accumulates and occurs over time, and the algorithm is applied to a practical situation where noise is not in ideal Gaussian distribution.

Description

technical field [0001] The present invention solves the problem of inaccurate battery estimation of pure electric vehicles, and thus proposes a method for estimating the remaining power of the battery with strong tracking and filtering based on noise interference. It is specially designed to the battery equivalent first-order RC model, noise interference filtering, and the battery estimation method of the CKF (Cubature Kalman Filter) volumetric Kalman filter algorithm with strong tracking. Background technique [0002] With the depletion of traditional petrochemical energy, our country is vigorously promoting new energy. Among them, the application and popularization of new energy vehicles are increasingly entering major cities and our lives. Electric vehicle management system (Battery Management System) is one of the core technologies in new energy vehicles. Accurate estimation of the remaining battery power is also one of the difficulties for major companies to overcome. ...

AI Technical Summary

Problems solved by technology

[0007] The problem solved by the present invention is to provide a battery estimation method based on strong tracking volume Kalman filter based on noise interference, to estimate the remaining power of the battery, and to solve the problem that the performance of the battery pack system is affected by its own internal factors (such as the rate of charging and discharging and internal Fever) interference, while solving the problem of environmental noise interference in the use of the car, this method uses the strong tracking volumetric Kalman filter method as a tool, with the adaptive square root volumetric Kalman filter as an auxiliary, in the case of noise interference , use the adaptive square root volumetric Kalman filter algorithm to effectively filter out the signal of noise divergence, and directly use the strong tracking volumetric Kalman filter method to estimate the SOC of the battery when there is no noise interference

Images