A battery digital twin modeling method and system based on edge-cloud collaboration

By employing an edge-cloud collaborative battery digital twin modeling method, combined with a multiphysics coupling model and a compensation model, and utilizing knowledge distillation and dual timescale correction techniques, the real-time and accuracy issues of battery state prediction are resolved. This enables efficient and accurate prediction of battery state, supporting the safety and lifespan extension of the battery management system.

CN122113624APending Publication Date: 2026-05-29SHANDONG UNIV
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
SHANDONG UNIV
Filing Date
2026-02-14
Publication Date
2026-05-29

AI Technical Summary

Technical Problem

Existing battery state prediction methods struggle to balance real-time performance and accuracy. Mechanistic models are computationally complex and prone to parameter drift, while data-driven models rely on large amounts of data and are difficult to deploy, failing to meet the real-time and accuracy requirements of battery management systems.

Method used

We adopt a battery digital twin modeling method based on edge-cloud collaboration to construct a three-layer collaborative architecture of "edge-cloud". By combining a multi-physics coupling model and a compensation model, we use knowledge distillation technology to lighten the model and achieve continuous optimization of the model through dual time scale parameter correction and neural network correction, forming a closed-loop optimization mechanism.

Benefits of technology

It achieves high-precision real-time prediction of battery status, meets real-time requirements, improves prediction accuracy and adaptability, and supports battery safety management.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122113624A_ABST
    Figure CN122113624A_ABST
Patent Text Reader

Abstract

The application belongs to the technical field of battery digital twin modeling, and provides a battery digital twin modeling method and system based on end-edge-cloud cooperation, constructs a battery digital twin model, extracts key features from the battery digital twin model through a knowledge distillation technology and distills a light-weight battery digital twin model, deploys the battery digital twin model on a cloud side, deploys the light-weight battery digital twin model on an edge side, and performs online correction and iterative optimization on the battery digital twin model on the cloud side; the battery digital twin model parameters on the cloud side after correction and optimization are directly issued to the corresponding light-weight battery digital twin model on the edge side; according to real-time running data of the battery obtained on the end side, the light-weight battery digital twin model on the edge side after parameter updating is used for prediction, and multivariate prediction values of battery voltage, surface temperature and state of charge are obtained. The application realizes high-precision real-time prediction of multivariate states of the battery.
Need to check novelty before this filing date? Find Prior Art