Novel lithium battery SOC estimation method based on fractional order model

Abstract

The invention discloses a novel lithium battery SOC estimation method based on a fractional order model. The method includes: building a simplified fractional order battery model under direct-current excitation on the basis of electrochemical theory; establishing a testing scheme to collect pulse discharging data of a battery under different SOC; recognizing model parameters under different SOC on the basis of a lease square algorithm; building a relation table between the parameters and the SOC after studying change rules of the model parameters and SOC and finding that the model parameters and the battery SOC present obvious monotonic linear tendency; adopting fixed-point theorem to perform iterative calculation for several times on the basis of the relation table between the parameters and the SOC to enable the parameters to gradually converge, and finally acquiring SOC value which is more accurate and stable. The method is a new exploration in estimating battery SOC by building a mathematic function relation between the model parameters and the SOC.

Description

technical field [0001] The invention relates to a battery management system, in particular to a method for estimating the state of charge (SOC) of a lithium battery. Background technique [0002] Lithium-ion batteries are an important source of power for electric vehicles, and their performance and status play a decisive role in the entire vehicle. The battery state of charge (SOC) indicates the remaining capacity of the battery, which is of great significance for electric vehicles. On the one hand, the battery management system controls battery charge and discharge according to the SOC value, which can protect the battery from overcharge and discharge, prolong life and reduce costs. On the other hand, electric vehicles identify the control strategy of the vehicle based on the SOC value of the power battery, which has a lot to do with its energy distribution. If a more accurate state of charge of the battery can be obtained, the range of the SOC can be expanded, thereby im...

AI Technical Summary

Problems solved by technology

[0003] The inside of the battery is a highly nonlinear electrochemical system. It is difficult to model such a system, and the SOC value of the battery cannot be directly measured.

The ampere-hour method is often used to estimate the SOC of the battery. The method is simple and easy to implement, but the cumulative error in the calculation process will continue to increase with the use of the battery, and its compensation coefficient is difficult to estimate; the open-circuit voltage method can estimate the SOC more accurately but requires Longer standing time stabilizes the electrochemical reaction inside the battery; the parallel structure and learning ability of fuzzy reasoning and neural network can realize accurate estimation of SOC, but a large amount of sample data is required for training, and the estimation accuracy is affected by training data and training methods Relatively large; electrochemical impedance spectroscopy (EIS) is often used to test the frequency domain response of the battery, but research shows that the EIS method is too complicated and difficult to update the battery SOC in real time; the Kalman filter algorithm is an optimal autoregressive data processing algorithm, It can realize the minimum variance estimation of the state of the system, and has great advantages in applications where the battery load fluctuates frequently and the operating current changes drastically.

When the algorithm is used, it is generally assumed that the noise is zero-mean white noise, and the variance of the noise is known, but in most cases the characteristics of the noise are unknown. At this time, the estimation performance of the algorithm decreases, and even causes divergence.

Images