Abstract
The working principle of fiber Bragg grating (FBG) sensors is mostly based on the tracking of the Bragg wavelength shift. To accomplish this task, different algorithms have been proposed, from conventional maximum and centroid detection algorithms to more recently-developed correlation-based techniques. Several studies regarding the performance of these algorithms have been conducted, but they did not take into account spectral distortions, which appear in many practical applications. This paper addresses this issue and analyzes the performance of four different wavelength tracking algorithms (maximum detection, centroid detection, cross-correlation and fast phase-correlation) when applied to distorted FBG spectra used for measuring dynamic loads. Both simulations and experiments are used for the analyses. The dynamic behavior of distorted FBG spectra is simulated using the transfer-matrix approach, and the amount of distortion of the spectra is quantified using dedicated distortion indices. The algorithms are compared in terms of achievable precision and accuracy. To corroborate the simulation results, experiments were conducted using three FBG sensors glued on a steel plate and subjected to a combination of transverse force and vibration loads. The analysis of the results showed that the fast phase-correlation algorithm guarantees the best combination of versatility, precision and accuracy.
Highlights
Fiber Bragg grating (FBG) sensors can be dated back to 1978, when Hill et al [1] discovered that refractive index variation can be formed in optical fibers
We present a survey on the performance of four peak detection techniques, maximum detection (MD), centroid detection (CD), CC and fast phase-correlation (FPC), when applied to dynamic measurements of distorted FBG spectra
We presented a comparison of four demodulation algorithms for dynamical measurements of distorted FBG spectra
Summary
Fiber Bragg grating (FBG) sensors can be dated back to 1978, when Hill et al [1] discovered that refractive index variation (i.e., gratings) can be formed in optical fibers. Can lead to non-uniform strain field distributions along the sensor grating When this happens, the original FBG reflected spectrum becomes distorted and shows multiple peaks, as well as more or less pronounced side lobes. The impacts of FBG spectral deformation on the interrogation performance has been investigated a few times These studies can be grouped into two main categories: the first focusing on the reconstruction of the non-uniform strain field in static or quasi-static condition [16,25] using neural networks and genetic algorithms; the second dealing with the development of new types of interrogator systems with higher performance in terms of accuracy and interrogation speed [26,27]. Performed dynamical strain measurements of embedded FBG sensors using a fast interrogator with full spectrum acquisition.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
