Lithium ion battery SOC (State Of Charge) prediction method based on big data and bp (Back Propagation) neural network

Abstract

The invention provides a lithium ion battery SOC (State Of Charge) prediction method based on big data and a bp (Back Propagation) neural network. The method comprises the following steps that: collecting battery external characteristic parameters to establish a dataset of battery SOC big data; establishing a training set and a test set; constructing multiple bp neural network prediction models; independently putting the dataset in the bp neural network prediction models of different parameters to obtain measurement accuracy; and according to measurement accuracy obtained by the bp neural network prediction models of different parameters, carrying out analysis to obtain prediction results. By use of the lithium ion battery SOC prediction method based on the big data and the SOC big data, through a SOC prediction big data set, the data can be conveniently effectively mined, and prediction accuracy is guaranteed; through the distributed bp neural network prediction models, the battery issubjected to SOC accurate prediction according to battery external parameters, and accuracy is high. Especially under big data, the battery external parameters continuously change. The method still can accurately predict the value of the battery SOC, is high in usability and can be widely applied in reality.

Description

technical field [0001] The present invention relates to the technical field of battery management, and more specifically, relates to a lithium-ion battery SOC prediction method based on big data and bp neural network. Background technique [0002] Due to the scarcity of energy and the increasingly serious environmental problems, electric vehicles have become a major trend to replace fossil fuel vehicles. Lithium-ion batteries are the core power components of new energy vehicles. In order to ensure the good performance of the battery and prolong its service life, it is necessary to manage the battery effectively. The premise is that the state of charge (SOC) of the battery must be known accurately and reliably. SOC is an internal characteristic of the battery that cannot be directly measured, but can only be obtained by predicting some directly measured external characteristic parameters such as its voltage, current, temperature, internal resistance, and capacitance. [000...

AI Technical Summary

Problems solved by technology

[0005] The present invention provides a lithium-ion battery SOC prediction based on big data and bp neural network in order to overcome the technical defect that all parameters need to be used for prediction when the existing battery SOC prediction method is applied, and the actual application cannot be well obtained. method

Images