Industrial equipment strain data multichannel acquisition system based on fiber grating sensing
By constructing a three-dimensional state vector and a Gaussian mixture model, and dynamically adjusting the exposure level and threshold, the demodulation failure problem caused by the difference in channel reflection spectrum intensity in the fiber Bragg grating sensing system was solved, achieving high accuracy and reliability of strain data.
Patent Information
- Application Number
- CN202610489468.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-04-14
- Publication Date
- 2026-07-17
AI Technical Summary
Existing fiber Bragg grating sensing systems in industrial equipment suffer from significant differences in reflectance spectrum intensity due to variations in the environment, fiber loss, and light source coupling efficiency at different measurement points. Unifying exposure time and fixing thresholds makes it difficult to meet the peak-finding requirements of both strong and weak channels, leading to demodulation failures or increased errors in some channels. Furthermore, the lack of dynamic identification and adaptive parameter adjustment mechanisms affects the continuity and reliability of strain data.
The system employs a baseline construction module, a state determination module, an adaptive demodulation module, and an iterative correction module. By constructing a three-dimensional state vector and a Gaussian mixture model, it dynamically adjusts the exposure level and threshold. Combined with the CA-CFAR algorithm and the Kalman prediction model, it achieves channel-level adaptive demodulation and parameter correction.
It significantly improves the success rate of multi-channel demodulation, ensures that reflection peaks can be clearly identified in both strong and weak channels, improves the accuracy of strain calculation, enhances the system's robustness to channel characteristic drift, and prevents long-term performance degradation.