The traditional temperature measurement system used the ratio of stokes light and anti-stokes light to demodulate the temperature, and there was an error when the ratio used directly. The propagation speed of stokes light and anti-stokes light in the same fiber was different because of dispersion effect. The two kinds of reflected light were obtained by photoelectric converter at different time when the scattering occurred at the same place. Therefore, there was a time warping between the two signals after data acquisition. This phenomenon would lead to measurement error in the process of demodulating temperature. A new method of error correction for the distributed temperature measurement system was used to eliminate the signal dislocation by correcting the stokes signal. In the absence of correction, the antistokes signal peak and the stokes signal peak occurred in different place, which leaded to errors in the process of demodulating temperature. This correcting method was used. The demodulation error was reduced from 2.3 °C to 1.2 °C. The shortest time was 4.96s when detected temperature rise rate was not less than 5 °C/min in the case of 10km fiber carried by this system. It was showed that this technique could improve the performance of the traditional distributed fiber Raman temperature measurement system, and it would have a good aspect.
In order to monitor the early fire situation caused by coal spontaneous combustion or cable aging, cable heated, cable overcurrent and other reasons, a dual channel and fast response temperature detection system is designed based on the Raman scattering principle in which anti-stokes signal in back Raman scattering is sensitive to temperature. The performance of the temperature detection system is obtained through experiments. The experimental data shows that the maximum error of temperature measurement is 0.7°C, the stability is 1°C in 30 minutes when both channels carry 2.5km sensing fiber, single channel response time is less than 17s and position-calculated error is less than 2m. This system would have a good application prospect in distributed fire detection and early warning under mine.
In distributed optical fiber sensing system of BOTDR, the frequency of the backward Brillouin scattering signal is modulated by temperature and strain, so the frequency of the signal can be attained with optical fiber to realize the temperature and strain measurement in remote. While, the scattering signal is very weak, the noise is large, and have a frequency width of several decades megahertz. In tradition, the complete high-frequency Brillouin scattering signal is obtained by adopting the method of frequency scanning which capture the frequency in section through changing the rate of sweep frequency module. And then, using multiple averaging measurement to reduce noise. But it is difficult to acquire the signal rapidly and eliminate the interference of noise. Therefore, an edge filter approach is proposed in this paper to get the whole scattering signal, which convert the frequency information into energy message. In order to better represent the effectiveness of this method, an experiment were taken. And the result shows that: SNR had been greatly enhanced, sampling time was reduced to the cost of getting one frequency point when using frequency scanning. It demonstrate that the proposed method can collect the signal quickly and be beneficial to demodulate temperature and pressure in time.