Parameter estimation OCV based full temperature SOC estimation method

Abstract

The invention relates to a parameter estimation OCV (Open Circuit Voltage) based full temperature SOC (State of Charge) estimation method and belongs to the technical field of battery management. Themethod includes the following steps: S1, selecting a power battery to be tested and a required equivalent circuit model, and determining a system state and model parameters that need to be identifiedon-line; S2, constructing an experiment and carrying out a variable power cycle experiment to the power battery, and then recording experiment data; S3, identifying a battery state and the model parameters online on the basis of the recorded operating condition experiment data and an adaptive joint extended Kalman filter algorithm; and S4, establishing an OCV-SOC-T model based on parameter estimation, and realizing SOC accurate estimation of the power battery in the full temperature range on the basis of the model and experiment operating conditions. The model adopted by the invention is simple, the algorithm is not complicated, highly robust on-line estimation of the system state of the battery and the model parameters can be realized, and thus parameter estimation OCV based full temperature SOC accurate estimation is realized.

Description

technical field [0001] The invention belongs to the technical field of battery management, and relates to a full-temperature SOC estimation method based on parameter estimation OCV. Background technique [0002] The accurate estimation of battery SOC is the basis of the battery management system, which is directly related to the accurate estimation of other battery states such as the power state, temperature state, and health state of the battery. [0003] At present, research institutions at home and abroad have relatively matured the research on SOC estimation of power batteries. The SOC estimation methods currently proposed can be roughly divided into five categories: model-free SOC estimation methods, SOC estimation methods based on surface characteristic mapping models, and black box-based SOC estimation methods. Model-based SOC estimation methods, electrochemical model-based SOC estimation methods, and equivalent circuit model-based SOC estimation methods. Model-free ...

AI Technical Summary

Problems solved by technology

Model-free SOC estimation includes two basic methods, namely the ampere-hour integration method and the discharge experiment method. The former method is simple and stable, and is currently widely used in engineering, but its estimation accuracy depends on the accuracy of the current sensor and the initial SOC. The latter is only suitable for laboratory measurement; the SOC estimation method based on the surface characteristic mapping model includes the working voltage method, the open circuit voltage method, and the DC impedance and AC impedance methods. The former two are easy to establish the mapping relationship between voltage and SOC, but they are not Considering factors such as aging and voltage hysteresis, the SOC estimation accuracy will gradually decrease over time, and the latter two are difficult to be applied to electric vehicles; SOC estimation methods based on black-box models include fuzzy logic, artificial neural networks, and support vectors. This type of method is convenient for modeling, but it requires a large amount of training data and a large amount of calculation, so it is difficult to apply it to the vehicle BMS; the SOC estimation method based on the electrochemical model has a complex model, and it is difficult to accurately identify many parameters. It tends to be used for battery characteristic analysis; the SOC estimation method based on the equivalent circuit model can use various simplified equivalent circuit models combined with the optimal estimation algorithm to achieve more accurate SOC estimation. This type of method has a simple model and a small amount of calculation. It is expected to be used in vehicle BMS

[0004] In the SOC estimation method based on the equivalent circuit model, the current research rarely considers the hysteresis effect of the open circuit voltage of the battery during charging and discharging, and how to effectively control the noise interference in data measurement and dynamic estimation has not been well understood. good solution

Images