Battery set service life prediction method based on battery car state data

Abstract

The invention relates to a battery set service life prediction method based on probability statistics according to user use habits of different types of battery cars. The method comprises steps that step 1), statistics and analysis on user habits influencing battery car battery service life are carried out, and two important factors include mileage and the battery discharging time; step 2), classification of battery car types is carried out according to voltage values, and a battery car user use habit and battery service association model is established; step 3), parameter estimation is carried out for the association model through utilizing a training sample set; and step 4), to-be-tested new user mileage and the accumulative discharging time are substituted to a corresponding battery service life state function for calculation, a standard quantity of the battery service lift state is acquired, through comparison with a set threshold, a battery service life state conclusion of a present battery car user is acquired.

Description

technical field [0001] The invention belongs to the technical field of life prediction of power battery packs, and the method is particularly suitable for detecting and diagnosing the service life status of the battery packs of different types of battery vehicles under the usage habits of users. Background technique [0002] In recent years, convenient, energy-saving and environmentally friendly battery cars have become a green means of transportation favored by urban residents. At present, the batteries commonly used by domestic battery car manufacturers mainly include lead-acid batteries, nickel metal batteries, and lithium-ion batteries according to chemical materials. These batteries of different materials are the most vulnerable parts of electric vehicles, and have been facing constraints such as not being durable and having a short lifespan. In addition, due to the large differences in road conditions and behavior habits of users during driving, it is difficult to ach...

AI Technical Summary

Problems solved by technology

However, the model is very effective in predicting the capacity in the early stage of the battery charge and discharge cycle, but the effect in the later stage is not very good.

[0007] To sum up, the existing power battery life prediction models are usually based on electrochemical detection or equivalent circuit models based on standard working conditions. These methods only target the internal characteristic parameters of the battery without considering the actual working conditions of the battery car. Research shows that improper user behavior will lead to lower battery pack utilization and shortened service life.

At the same time, none of the existing prediction methods collect and analyze a large amount of historical data, so it is impossible to find out the aging law objectively, which increases the error

Images