Optical strain gauge signal processing

Abstract

(Fiber Bragg Grating (FBG) sensors are becoming more popular in the sensing applications where high precision is required. Fiber Bragg Grating is a periodic variation of refractive index in an optic fiber. Basically FBG is a wavelength selective filter for the optical signals. Filter frequency is based on the pitch length of the grating .Due to strain, the pitch length changes, which in turn changes the peak wavelength of the reflected spectrum. From the center wavelength shift, strain can be computed. Due to temperature, there is an indirect strain inducement in the fiber which can alter the peak wavelength in an indirect way .By computing the change in peak wavelength shift, temperature can also be measured .In this work, noise filtering was carried out with different filters like Butterworth LPF, Gaussian Filter and wavelet de-noising technique. After each filtering, Gaussian fitting technique was used to calculate the peak wavelength of the reflected spectrum since the reflected spectrum resembles with Gaussian distribution. It was observed that Wavelet de-noising associated with Gaussian fitting is more better in terms of peak wavelength shift after filtering .Studied the effect of temperature by applying constant strain at different temperatures in a digitally controlled oven .Observed that peak wavelength is same for a particular temperature at different instances while thermal cycling was carried out . A bonded sensor was tested to find the correlation between signals at different temperatures and found that the signal spectrum is highly correlated at different temperatures and thereby came to a conclusion that existing wavelet de-noising technique with Gaussian fitting can be used reliably for measuring the central wavelength shift.