A method and system for hierarchical early warning and linkage control of thermal runaway of energy storage batteries based on multi-dimensional parameter fusion prediction

By employing multi-dimensional parameter fusion prediction and scenario-adaptive safety control technology, the problems of insensitive early feature capture and delayed fire suppression in lithium-ion battery thermal runaway early warning have been solved. This enables extremely early warning and precise control of lithium-ion battery thermal runaway, improving the safety and management efficiency of energy storage batteries.

CN122432866APending Publication Date: 2026-07-21NANJING ZIWEIYUAN SCIENCE & TECHNOLOGY RESEARCH CO LTD +1
View PDF 4 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
NANJING ZIWEIYUAN SCIENCE & TECHNOLOGY RESEARCH CO LTD
Filing Date
2026-04-16
Publication Date
2026-07-21

AI Technical Summary

Technical Problem

Existing technologies for early warning of thermal runaway in lithium-ion batteries suffer from problems such as insensitive early feature capture, high false alarm rate, and ineffective suppression of reignition by fire extinguishing media. In particular, in large-scale energy storage chambers, there is a lack of a linkage control scheme that deeply integrates BMS electrical data, environmental chemical data, and physical thermal field data, resulting in delayed fire extinguishing or water damage and pollution in undamaged areas.

Method used

A multi-dimensional parameter fusion prediction method is adopted. By collecting multi-dimensional characteristic parameters of energy storage batteries, time series learning is performed using prediction models, and early risk prediction values ​​are generated by combining gradient boosting decision technology. The weights are adjusted based on scenario adaptive safety control technology to construct a thermal runaway trend index and match hierarchical prevention and control strategies for coordinated control.

Benefits of technology

It enables very early warning of thermal runaway in lithium-ion batteries, improves the accuracy and timeliness of warnings, reduces safety hazards, optimizes the design and control strategies of energy storage batteries, and constructs a closed-loop safety management system for the entire process.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122432866A_ABST
    Figure CN122432866A_ABST
Patent Text Reader

Abstract

The application relates to the technical field of electrochemical energy storage safety monitoring, and discloses a thermal runaway grading early warning and linkage control method and system for energy storage batteries based on multi-dimensional parameter fusion prediction, which comprises the following steps: performing time sequence learning on multi-dimensional characteristic parameters, combining gradient boosting decision technology, and generating a multi-dimensional characteristic early risk prediction value; dynamically adjusting the dynamic weight of the multi-dimensional characteristic early risk prediction value according to different scenes and risk assessment levels, generating a thermal runaway trend index, and combining the multi-dimensional characteristic early risk prediction value and an energy storage battery voxel space model to evaluate the performance state of the energy storage battery; and taking the thermal runaway trend index and the performance state of the energy storage battery as the basis, matching multi-dimensional data early warning risk rules to output a grading prevention and control strategy, and using the grading prevention and control strategy to control and process the thermal runaway of the energy storage battery. The application can not only describe the propagation trend and risk area positioning of thermal runaway in a battery pack, but also provide data reference for the design optimization of energy storage cell spacing.
Need to check novelty before this filing date? Find Prior Art