| Distributed optical fiber sensors possess the features of small in size,low in weight, immune to electromagnetic interference and capable of remote measurement. They have been widely employed for the detection of the temperature and strain in various applications such as the forest fire alarm, the intrusion detection of oil pipeline and the structure health monitoring, etc. Compared to the distrituted optical fiber sensing technologies based on Rayleigh scattering and Raman scattering, the technology called Brillouin optical time domain analysis (BOTDA) based on stimulated Brillouin scattering (SBS), which has been one of the hot research topics in the field of optical fiber sensor for several years, has the key advantage of the better performance for long distance sensing.The performance of a BOTDA sensor is mainly determined by four parameter specifications, which are spatial resolution, frequency accuracy,measurement range and measurement time. In this dissertation, the physical limitations of these four parameter specifications have been deeply analysed and evaluated, and several innovative technologies have been proposed to overcome the limitaions and to improve the performance of the sensors.The main innovative research efforts and achievements are summarized as follows:1) Although the BOTDA sensors based on dual-frequency probe wave have been widely used to overcome the non-local effect, the investigation in this dissertation shows that the Brillouin gain/loss spectrum (BGS/BLS) would be highly distorted when the power of the probe wave is higher than -6 dBm, further resulting in the error on estimated Brillouin frequency shift (BFS). In this dissertation, a novel BOTDA sensor based on fixed dual-frequency probe wave is proposed, in which the the synthetic effect attributed to the Brillouin gain spectrum generated by the upper-frequency probe component and the Brillouin loss spectrum generated by the lower-frequency probe component can be cancelled out. As a result, the BGS/BLS distortion problem existing in conventional BOTDA scheme can be overcome. The experimental results indicate that the proposal allows the probe power being increased from -6 dBm to the SBS threshold (5 dBm) without BGS/BLS distortion. The sensing range of more than 100 km is achieved without any advanced method to improve SNR such as pulse coding and Raman distributed amplification.2) The detrimental impact of the second echo phenomenon that commonly exists in pre-pumping BOTDA methods (including bright pulse method, dark pulse method and ? phase shift pulse) is thoroughly investigated by modeling and solving the instantaneous Brillouin three-wave coupled equations. The presented analysis points out that the second echo phenomenon would lead to systematic error on the estimated Brillouin frequency shift, the maximum of which could reach up to 8.5 MHz. Instead of using conventional three-section pulse as pump wave, a novel four-section pulse is proposed, in which an additional pulse section with parabolic-amplitude distribution is employed. The key parameters of the proposed pulse are optimized by combining the analytical solution of the Brillouin equation and the iterative algorithm.The theoretical analysis and experimental results demonstrate that the proposal can realize sub-meter spatial resolution without the impact of the second echo, and therefore, the BFS change is correctly achieved. Besides, a novel BOTDA method based on long pump pulse is also proposed for eliminating the impact of the second echo. The long pump pulse consists of two parts, in which the width of the sensing part is exactly equal to twice of the fiber length, while the width of the other part, which is used for pre-exciting the steady state acoustic wave, should be longer than 20 ns. The distrituted temperature or strain information is obtained by using logarithmic and differential algorithom. The spatial resolution of 3 cm is achieved in the experimental demonstration without the impact of the second echo.3) The capacity of realizing "gain process" and "loss process"simultaneously is the unique bipolar feature of SBS process, which can be used to design the BOTDA sensor with higher performance. For instance,the dual-frequency differential pulse pair BOTDA method can achieve sub-meter spatial resolution without using extra measurement time, and the bipolar Golay coding BOTDA method can provide higher SNR improvement than the methods using unipolor code. However, this dissertation points out that those two methods based on single-tone probe wave commonly suffer from the pump depletion problem, which would give rise to the unbalance of the power of the two frequency components on pump branch as they propagate in the fiber. This leads to the deterioration of the spatial resolution in dual-frequency differential pulse pair BOTDA method, and BFS error in bipolar Golay coding BOTDA method. Instead of using single-tone probe wave, a novel concept of three-tone probe wave based BOTDA sensor is proposed to dynamically compensate the pump depletion phenomenon. The theoretical analysis and the experimental results demonstrate that the delayed dual-frequency differential pulse pair BOTDA method with three-tone probe can at least achieve 50 cm spatial resolution and 25 km sensing range. The bipolar Golay coding BOTDA method with three-tone probe reaches a sensing range of 100 km in a 200 km-long fiber-loop, with 2 m spatial resolution and 0.9 MHz frequency uncertainty on the BFS when using 1000 averages per trace. The improvement brought by the proposed method and its implementation is quantified by a figure-of-merit (FoM) of 380000, which constitutes the highest FoM so far reported in a Brillouin distributed fiber sensor. |
PDF Full Text Request