Fastening abnormality determination device and vehicle
By generating a learned model and using current waveform data to determine the loosening state of the fastened structure, the problem of misjudgment in the existing technology is solved, and high-precision fastening anomaly judgment is achieved.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- TOYOTA JIDOSHA KK
- Filing Date
- 2025-11-25
- Publication Date
- 2026-07-17
AI Technical Summary
Existing technologies are prone to misjudgment when determining whether a fastener is loose, especially when the current value control method changes, making it difficult to accurately identify the looseness of the fastener.
A machine learning method is used to generate a fully learned model. Current waveform data is acquired through a current sensor. The first and second fully learned models are used to determine whether the fastened structure is loose during the current convergence period and the target value stabilization period, respectively, to suppress misjudgments caused by current fluctuations and instantaneous abnormal values.
It enables high-precision determination of whether a fastened structure is loose under different conditions, reducing the occurrence of false judgments and improving the accuracy of fastening anomaly determination.
Smart Images
Figure CN122402304A_ABST
Abstract
Citation Information
Patent Citations
Power storage system
JP2019132608A