滚动轴承故障诊断方法、计算机设备及计算机程序产品
By constructing a dual-path parallel hybrid model driven by impact features, the problems of insufficient feature extraction of rolling bearing vibration signals and insufficient stability of multi-category fault diagnosis are solved, and high-precision and high-robust fault diagnosis effect is achieved.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- SHENZHEN UNIV
- Filing Date
- 2026-06-17
- Publication Date
- 2026-07-17
AI Technical Summary
Existing technologies are insufficient in extracting vibration signal features of rolling bearings under complex working conditions, making it difficult to coordinate the modeling of temporal and global features. They also lack stability in multi-category fault diagnosis, and traditional methods rely on human experience and are susceptible to noise interference.
A dual-path parallel hybrid model driven by impact features is constructed, including time-frequency local information extraction, feature sequence combination, long short-term memory network modeling, Transformer branch modeling, and impact-driven gating fusion. The model is trained by time-frequency features and jointly optimized by gating constraints.
It achieves high-precision and robust diagnosis of rolling bearing faults under complex working conditions, making full use of multi-scale structural information of vibration signals to enhance the accuracy and stability of fault diagnosis.
Smart Images
Figure CN122409196A_ABST