一种融合多模态时序数据的风机故障预警与诊断方法
By using a three-state polarity synchronous counting mechanism to process the vibration, current, and temperature data of the wind turbine in parallel, the problem of information loss during multimodal data fusion in existing technologies is solved, enabling effective identification and differentiation of early faults and improving the accuracy and self-optimization capability of the early warning system.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- CHINA POWER INVESTMENT NORTHEAST NEW ENERGY DEV CO LTD
- Filing Date
- 2025-09-29
- Publication Date
- 2026-07-17
AI Technical Summary
Existing technologies, when fusing multimodal wind turbine data, cannot effectively capture early signs of faults due to information loss, and have difficulty distinguishing between real faults and external interference, resulting in the early warning system lacking the ability to identify early faults.
A three-state polarity synchronous counting mechanism is adopted. By processing the vibration, current and temperature time series data of the fan in parallel, the disorder degree, deviation degree and temperature change trend characteristic sequence are calculated. Combined with spectrum symmetry verification and chronic fault diagnosis, a fault warning signal is generated, and active lubrication status detection and feedback are performed under low load conditions.
It enables early warning of the disordered process of multimodal data collaboration, can identify the evolution of the system from a healthy state to an abnormal state, distinguish between internal mechanical failures and external environmental interference, and improves the accuracy and self-optimization capability of the early warning system.
Smart Images
Figure CN121139293B_ABST