Self-adaptive electric vehicle SOC estimation method based on big data

Abstract

The invention discloses a self-adaptive electric vehicle SOC estimation method based on big data, and the method comprises the steps: the time, longitude, latitude, SOC value, vehicle speed, odometertotal mileage value, battery pack output total current and battery pack output total voltage of a vehicle are collected from the vehicle in advance, and serve as a training data set; acceleration, distance and energy consumption values are calculated according to time, speed, longitude and latitude, current and voltage values in a training data set, characteristic speed, acceleration, distance anddependent variable energy consumption values are used for constructing an extreme random decision tree model, and an SOC prediction model based on mileage and energy consumption is obtained accordingto a total mileage value, an energy consumption value and an SOC value of an odometer. Therefore, the final SOC prediction model is obtained by the SOC prediction model based on mileage and energy consumption according to the genetic algorithm, the model can update the data in the training data set every other T time, and the self-adaptive prediction effect is achieved.

Description

technical field [0001] The invention relates to the field of electric vehicle SOC estimation, in particular to an adaptive electric vehicle SOC estimation method based on big data. Background technique [0002] In recent years, with the rapid development of lithium-ion battery technology, the status of electric vehicles is increasing day by day. However, due to the obvious nonlinear and time-varying characteristics of the state of charge (SOC) of lithium-ion batteries, the prediction of SOC has always been a key and difficult point in the field of electric vehicles. Therefore, the driver's mileage anxiety phenomenon often occurs. And the long-term SOC prediction is also of great significance to intelligent transportation, unmanned driving and other aspects. [0003] At present, there are mainly three methods commonly used to estimate SOC: the first is the estimation method based on the battery model represented by the ampere integral method, the open circuit voltage method...

AI Technical Summary

Problems solved by technology

However, due to the obvious nonlinear and time-varying characteristics of the state of charge (SOC) of lithium-ion batteries, the prediction of SOC has always been a key and difficult point in the field of electric vehicles. Therefore, the driver's mileage anxiety often occurs

When using the first estimation method based entirely on the battery model, the error of this type of method will gradually accumulate as the forecast time span grows, so the forecast result may have a large error; the recursive algorithm relies on the battery model, And with the increase of the single prediction time span, the prediction accuracy rate drops significantly, and the uncertainty of the recursive algorithm may continue to accumulate during the calculation process, which may seriously affect the results and even cause the algorithm to diverge; the third type of machine learning method is insufficient The advantage is that a large amount of data support is required, the amount of calculation is large, and the algorithm model is not easy to be trained

Images