The quality of the reflection spectrum of the Fiber Bragg Grating (FBG) will directly affect the performance of the grating sensor network. Aiming at the problem of quality detection of fiber Bragg grating sensor network prepared online, this paper proposes a reflection spectrum quality identification algorithm based on Support Vector Machine (SVM). Firstly, the characteristic information of the original reflection spectrum is extracted based on theory and simulation. Then the raw feature information is preprocessed using a "signal segmentation and search model" to find the approximate locations of the original reflection spectrum. Finally, the kernel function is used to optimize the SVM network model until the recognition accuracy of the reflection spectrum quality feature meets the requirements. 2000 groups of reflection spectrum characteristic data were put into SVM recognition network for training, the recognition effect reached 99.9%. Finally, the non-uniform temperature field is established, and the experimental results show that the algorithm can identify the distortion spectrum pattern of the fiber Bragg grating array with more than 99% accuracy. It effectively improves the recognition accuracy of reflection spectrum peak sag and distortion and provides a new idea for the quality detection of core components of fiber Bragg grating sensor network.
We propose an improved Gaussian curve fitting method based on the Hilbert transformation (HTG) to tackle the ineffectiveness of the traditional peak-seeking algorithm in detecting the multi-peak Fiber Bragg grating (FBG) reflection spectra. A five-point sliding filter is used to process the FBG reflection spectral signal, de-noise and smooth the noise, and select the optimal threshold point by the Hilbert transformation (HT). The sub-spectra of the multiple FBG reflection spectral signals were derived and the initial positioning of the spectral peaks were achieved. The Levenberg–Marquardt (LM) algorithm is used to extract the Bragg wavelength from the segmented sub-spectral signals as well as optimize the Gaussian curve fitting coefficients. The HTG-LM algorithm is then proposed, and is optimized and utilized to achieve precise positioning of the spectral peaks. The theoretical analysis and experimental results showed that the proposed HTG-LM algorithm could dynamically detect the multiple reflection spectra of the FBG sensing system with good stability, and at the same time, reduce the amount of peak-seeking data, which is highly beneficial to improve the signal demodulation rate. The peak detection accuracy of the proposed algorithm is better than 1 pm and the precision is better than 4 × 10 − 7 pm, which indicates that this HTG-LM algorithm provides an accurate demodulation algorithm for the FBG sensor networks. As a result, it is a promising multi-peak detection algorithm proposed by this paper to be applied to the FBG sensing systems.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.