SOC estimation method for lithium battery based on state transition optimized RBF neural network

Abstract

The invention discloses a SOC estimation method for a lithium battery based on a state transition optimized RBF neural network. The method relates to the technical field of electric automobiles, wherein the method comprises the steps that: (1) offline training sample data are collected, normalization process is conducted on all training samples; (2) a SOC estimation model for a lithium battery based on a RBF neural network is established; (3) a STA optimization algorithm is adopted to optimize the established RBF neural network model; (4) the trained RBF neural network and each parameter are saved, the trained RBF network is used for conducting estimation on the SOC of a lithium iron phosphate battery; by means of the method, the SOC of the lithium battery is accurately estimated, the method has the advantages of high estimation precision, strong reliability, simple estimation model and the like, which can be widely applied in the technical field of power battery of the electric automobiles.

Description

Technical field [0001] The invention relates to the technical field of electric vehicle power batteries, in particular to a method for estimating the SOC of a lithium battery based on a state transfer optimized RBF neural network. Background technique [0002] With the deterioration of the global environment and the serious shortage of resources, especially the aggravation of domestic PM2.5 pollution in recent years, the traditional automobile industry at home and abroad has undergone tremendous changes, and the development of new energy vehicles has become the focus of everyone's increasing attention; Among the many alternative energy sources for automobiles, electric energy is characterized by its safety, efficiency, and cleanliness, making new energy vehicles with power batteries as the main power source or auxiliary power source the main research object; the battery management system (BMS) is based on optimization The control system for managing and protecting the power ...

AI Technical Summary

Problems solved by technology

[0003] At present, the commonly used SOC estimation methods are: open circuit voltage method, impedance analysis method, ampere-hour measurement method, neural network method and Kalman filter method, etc.; the obvious disadvantage of the open circuit voltage method is that the battery needs to be left standing for a long time during the measurement, usually several times. One to more than ten hours, this method is only suitable for testing the SOC of the electric vehicle in the parking state; the impedance analysis method is to estimate the battery SOC by studying the relationship between the battery resistance and the SOC; but it is still very difficult to estimate the SOC by using the battery impedance , and the accuracy of the estimation cannot be guaranteed; the ampere-hour measurement method is the most used SOC estimation method in practice, and there will be some problems in the application of the ampere-hour measurement method; the error in the current measurement will lead to the calculation error of the SOC, the error Accumulate continuously and get bigger and bigger, and the error will be bigger under the condition of high temperature and current fluctuation; when using the Kalman filter method (KF, Kalman Filter) to estimate the SOC of the battery, the battery is regarded as a power system, and the SOC is As an internal state of the system; the Kalman filter algorithm used for SOC estimation mainly has the following two problems: First, it is highly dependent on the battery model. To obtain accurate SOC, it is necessary to establish a more accurate battery model, and the battery model’s The degree of accuracy is directly proportional to the complexity; the second is that the Kalman algorithm contains a large number of matrix operations, and the amount of calculation is large

Images