A deep learning-based device inspection data analysis method and system

By employing a deep learning-based equipment inspection data analysis method and a dual-path analysis framework, we can distinguish between equipment faults and environmental interference, generate multi-dimensional evaluation results, solve the problem of high false alarm rate in existing technologies, and improve the accuracy and safety of equipment fault identification.

CN122432916APending Publication Date: 2026-07-21SHANGHAI TIANZHOU SOFTWARE TECHNOLOGY CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
SHANGHAI TIANZHOU SOFTWARE TECHNOLOGY CO LTD
Filing Date
2026-04-25
Publication Date
2026-07-21

Smart Images

  • Figure CN122432916A_ABST
    Figure CN122432916A_ABST
Patent Text Reader

Abstract

The application relates to the technical field of equipment state monitoring, and discloses a device inspection data analysis method and system based on deep learning, which comprises the following steps: obtaining operation monitoring signals and environmental parameter information of a monitored device; obtaining characteristic parameters representing the state of the device according to the operation monitoring signals; comparing the characteristic parameters with pre-stored device fault characteristic modes to obtain a device probability index; combining the environmental parameter information, comparing the characteristic parameters with pre-stored environmental interference characteristic modes to obtain an environmental probability index; and generating a multi-dimensional evaluation result representing abnormal source distribution according to the device probability index and the environmental probability index. Through an innovative technical framework of'source separation evaluation and dynamic verification', the application solves the industry pain point that the root cause is difficult to accurately determine in device anomaly monitoring.
Need to check novelty before this filing date? Find Prior Art