An Adaptive Filtering Method Based on Parameter Estimation Criteria for SOC Estimation

Abstract

The invention belongs to the field of target tracking and discloses an adaptive filtering method based on parameter estimation criteria for SOC estimation. The invention proposes a parameter estimation criterion to estimate an accurate SOC estimation model, thereby transforming the adaptive filtering problem of SOC estimation into a constrained optimization problem. By solving the solution problem, the estimation of process noise covariance and observation noise covariance in SOC model is realized. The invention enables the SOC system to have the best estimation effect under the condition that the noise covariance is inaccurate.

Description

technical field [0001] The invention belongs to the field of state estimation, in particular to an adaptive filtering method based on parameter estimation criteria for SOC estimation. Background technique [0002] The State of Charge (SOC) of the battery refers to the amount of electricity stored in the battery at present, and is an important parameter of the control strategy of the battery management system, providing a judgment standard for the control strategy of the battery pack. At present, the commonly used SOC estimation methods include the ampere-hour integral method, the open circuit voltage method, the fuzzy neural network method and the Kalman filter method. The Kalman filtering method can obtain the minimum variance estimation of SOC by a recursive algorithm according to the collected voltage and current, and solve the problems of inaccurate initial SOC estimation and cumulative errors. [0003] Kalman filtering was first proposed by Kalman in 1960. It is a recu...

AI Technical Summary

Problems solved by technology

Subsequent researchers continue to improve the Sage-Husa adaptive filter, and they can only suppress the divergence of the filter when Q (or R) is known, and at the same time estimate R (or Q); the root cause is that the process noise When both covariance and observation noise covariance are unknown, it does not provide enough information to estimate both process noise covariance and observation noise covariance

Images