A fault diagnosis and early warning method for a power battery intelligent thermal management system
By constructing a dynamic spatiotemporal neural field model and a dual-channel verification mechanism, the micro-short circuit faults inside the power battery are identified, solving the problems of early warning lag and false alarms/missed alarms in the existing technology. This enables early fault identification and high-accuracy early warning of the power battery, improving the safety and reliability of the battery system.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- JIANGSU JIAHE THERMAL SYST RADIATOR
- Filing Date
- 2025-08-13
- Publication Date
- 2026-05-01
AI Technical Summary
Existing power battery thermal management systems are inadequate in identifying early and weak fault signals of thermal runaway caused by hidden defects such as micro-short circuits inside the power battery. They are unable to distinguish between true internal fault precursors and normal system dynamics or noise interference, resulting in delayed warnings and false alarms or missed alarms.
An adaptive graph neural network is used to construct a dynamic spatiotemporal neural field model. Combined with electrothermal delay analysis and a dual-channel verification mechanism, micro-short circuit signals are identified through cross-correlation delay detection and generative adversarial networks. Background noise signals are synthesized using generative adversarial networks for signal verification. Accurate early warning is achieved by combining a hierarchical response strategy.
It enables early identification and high-accuracy warning of micro-short circuit faults inside the power battery, eliminates the risk of misjudgment caused by cooling system fluctuations and sensor noise, and builds a rapid isolation closed-loop safety protection system from the risk initiation stage to the thermal runaway critical point, thereby improving the safety margin of the battery system.
Smart Images

Figure CN120816907B_ABST
Abstract
Citation Information
Patent Citations
Thermal runaway protection system and method for power battery
CN117962696A
Multi-dimensional battery thermal runaway judgment and early warning method, device, equipment and medium
CN119627276A