Battery SOC Estimation Method Based on Extended Kalman Filter with Small Model Error Criterion

Abstract

The invention discloses a method for battery SOC estimation based on small model error criterion expanding Kalman filter. The method comprises the steps of first establishing a variable-order RC model based on an AIC criterion and laying a good foundation for SOC precise estimation; obtaining data such as battery terminal voltage, current and corresponding model errors in an off-line mode under different working conditions and establishing a model error prediction model based on a fuzzy neural network; predicting model errors in an on-line mode based on the neural network in filtering, and performing measurement and updating on state estimation only when predicted errors are small so that the problem of filtering divergence caused by the model errors and system noise statistical characteristic uncertainty and the problem of SOC estimation fluctuation caused by battery terminal voltage jump can be solved. The method can effectively eliminate filtering estimation errors caused by the model errors and can be suitable for the dynamic process of a battery under various complex working conditions.

Description

technical field [0001] The invention relates to a battery SOC estimation method based on the small model error criterion extended Kalman filter. Background technique [0002] As a key component of electric vehicles, vehicle-mounted power batteries are crucial to the power, economy and safety of the vehicle, and are a key factor restricting the scale development of electric vehicles. In order to maximize the performance of the power battery and prolong the service life of the battery, it is very important to manage the battery effectively, and accurately obtaining the state of charge (SOC) of the battery is the core technology of battery management. Battery SOC estimation is an important basis for judging whether the battery is overcharged or overdischarged, and whether it needs to be balanced or replace a single battery. Therefore, improving the accuracy of battery SOC estimation is of great significance for improving battery efficiency, prolonging battery cycle life, and e...

AI Technical Summary

Problems solved by technology

Then, in the filtering, based on the neural network online prediction model error, the state estimation is measured and updated only when the predicted error is small, thus overcoming the problem of filtering divergence caused by the uncertain statistical characteristics of the model error and system noise

At the same time, the filter gain coefficient is introduced to optimize the filter gain matrix, which solves the problem of SOC estimation fluctuations caused by battery terminal voltage jumps

Images