Battery multi-model fusion modeling method based on self-adaptive weight method and battery management system

Abstract

The invention relates to a battery multi-model fusion modeling method based on a self-adaptive weight method and a battery management system. The self-adaptive weight method is used for calculating different models so as to switch model working modes based on different weights, it is ensured that output fusion models or selected models can work in an optimal mode under any usage condition, finally accurate and reliable prediction and tracking of key parameters and properties are achieved, and the overall quality of the control performance is improved. By deeply analyzing the precision, stability, calculated quantity, storage space and other property characteristics of different battery models under working conditions, multiple environments and multiple stages, online constantly updated multi-model fusion modeling is achieved, and the problem that accurate prediction of existing model properties is difficult to achieve for a battery system under internal and external factors such as full life cycle, uncertain and dynamic working condition and environments and single functions cannot meet all usage requirements is solved.

Description

[0001] Technical Field: The present invention relates to the technical field of power battery management, especially the field of power battery system modeling and state estimation. The present invention also relates to a battery management system. [0002] Background technique: [0003] Power batteries are widely used in the field of new energy vehicles. They are the core components of new energy vehicles. They are also a key factor in determining the service life, safety and economy of new energy vehicles. Power battery system is a kind of dynamic system with strong time-varying nonlinearity and non-uniformity. With the increasing attention to battery safety, thermal management and lifespan issues, the battery's charge, energy, health, safety, and peak value can be accurately obtained Status information such as power and prediction of real-time remaining life and remaining time are the core issues and the core functions of the battery management system that are concerned during ba...

AI Technical Summary

Problems solved by technology

The electrochemical principle model analyzes the performance change law of the battery during operation from the perspective of the essential electrochemical mechanism of the battery. This type of model has certain advantages in SOH estimation, RUL prediction, thermal monitoring, etc., but the model is extremely complex and has many parameters. The amount of calculation is large, and the performance of SOC estimation is not good at the same time

The equivalent circuit model analyzes a large amount of state data, builds a circuit model composed of electrical components based on the working principle of the system, and performs equivalent or approximate system dynamic characteristics. This type of model is mainly used for SOC estimation. The model is relatively simple and the calculation amount is small. , but it is difficult to achieve SOH estimation, RUL prediction and thermal monitoring, and it has a tendency to fail when the battery is in a low state of charge

The empirical model does not consider the electrochemical reaction and working process inside the lithium-ion battery, and directly mines the hidden battery state information and its evolution law from the battery performance test data and state monitoring data (voltage, current, temperature, impedance, etc.). The class model can overcome the problems of poor dynamic accuracy and poor universality of the model to a certain extent, but it requires a large amount of data for training, and is also susceptible to data uncertainty and incompleteness, and its robustness and adaptability are poor. Difference

It can be seen that although the relevant theories of various existing battery models are relatively mature, each model has its own advantages and disadvantages. All of them can only realize part of the functions of the battery management system, and perform poorly in other aspects or even cannot realize some functions at all. Or it fails under certain conditions. There is no battery model that can realize all the functions required by the battery management system under multiple working conditions, multiple environments, and multiple stages.

Images