设备异常状态识别方法、装置、存储介质及电子设备
By extracting power supply current, ADC sampling, attenuation value sequences, and electronic lock features from massive test logs, and combining statistical distribution and correlation models, hidden anomalies in SECC and EVCC products are identified, solving the problem of equipment being misjudged as qualified in existing technologies and achieving more accurate quality control.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- QIJING INFORMATION TECHNOLOGY (SHANGHAI) CO LTD
- Filing Date
- 2026-05-08
- Publication Date
- 2026-07-17
AI Technical Summary
Existing technologies cannot identify devices that are within the threshold range but have exhibited latent abnormalities such as parameter shifts, batch drift, multi-parameter coupling anomalies, and sequence fluctuation anomalies during SECC and EVCC product testing. This leads to devices with quality risks being mistakenly judged as qualified.
By extracting power supply current features, ADC sampling features, attenuation value sequences, and electronic lock features from massive test logs, and using statistical distributions based on historical normal devices to detect outliers, median deviations, abnormal fluctuations within sequences, and batch offsets, a correlation model between feature parameters is constructed, and a comprehensive defect score is calculated to mark potentially defective devices.
It can identify hidden anomalies that traditional methods cannot detect, reducing the chance of equipment being mistakenly judged as qualified and improving product reliability and quality control.
Smart Images
Figure CN122133043B_ABST