Bearing lubrication system fault diagnosis system based on intelligent sensing technology

By synchronizing signals with multimodal sensors and edge computing, and combining deep causal networks and knowledge graphs, the problems of multimodal signal timing misalignment and insufficient physical constraint verification in the fault diagnosis of bearing lubrication systems in existing technologies are solved, achieving efficient and accurate fault mode recognition and diagnosis.

CN122409191APending Publication Date: 2026-07-17SHANDONG XINSHENG AUTOMATION EQUIPMENT CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
SHANDONG XINSHENG AUTOMATION EQUIPMENT CO LTD
Filing Date
2026-04-23
Publication Date
2026-07-17

AI Technical Summary

Technical Problem

Existing fault diagnosis technologies for bearing lubrication systems rely on single or limited sensor signals, with signal processing concentrated in the cloud and lacking edge-end synchronous preprocessing capabilities. This results in timing misalignment of multimodal sensor signals, incomplete feature dimensions, an inability to quantify the intrinsic correlation between fault modes, and diagnostic results that are not verified by physical constraints, making them prone to misjudgment.

Method used

Multimodal sensors are used to collect vibration, acoustic emission, temperature and lubricating oil pressure signals. Edge computing nodes are synchronized and time-frequency domain features are extracted to construct a deep causal network model to quantify the causal contribution of fault modes. The model is combined with a knowledge graph for logical verification and causal path backtracking to eliminate fault entries that do not conform to physical constraints.

Benefits of technology

It achieves complete coverage of multi-dimensional features, quantifies the causal correlation of fault modes, eliminates fault modes with low correlation, ensures the physical consistency and accuracy of diagnostic results, reduces redundancy in remote signal transmission, and preserves the real-time performance and integrity of signals.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122409191A_ABST
    Figure CN122409191A_ABST
Patent Text Reader

Abstract

The application discloses a bearing lubrication system fault diagnosis system based on intelligent sensing technology, and relates to the technical field of intelligent diagnosis of industrial bearings, comprising a sensing collection module, a feature processing module, a cause-effect inference module, a candidate screening module and a graph verification module. The sensing collection module collects multi-modal signals of vibration, acoustic emission, temperature and lubricating oil pressure and transmits them to an edge computing node. The feature processing module completes signal synchronization alignment and joint feature extraction in time and frequency domains, and generates a multi-dimensional feature sequence. The cause-effect inference module constructs a deep cause-effect network model to calculate the cause-effect contribution degree of fault modes, and the candidate screening module screens according to a threshold value and generates a candidate fault set in combination with working conditions. The graph verification module eliminates illegal faults through logical verification and cause-effect backtracking. The system eliminates signal time sequence deviation, quantifies fault cause-effect correlation, guarantees that the diagnosis result is in line with physical constraints, and improves fault diagnosis accuracy and traceability effectiveness.
Need to check novelty before this filing date? Find Prior Art