Contribution to the advancement of Brillouin optical time-domain analysis sensors

Résumé

Distributed fiber optic sensors (DFOS) are becoming an increasingly used technology to monitor the integrity of structures. This is due to the fact that this technology can be embedded within the structure and provide distributed information of several relevant parameters for the structure, such as stress, temperature or strain. In DFOS the fiber itself is the transducer, and the measurement of a given parameter is provided continuously along the fiber at a particular spatial resolution, without blind spots. This is the main advantage of DFOS compared to other sensing technologies, the fact that DFOS provide information of a given parameter over thousands or hundreds of thousands of positions along the optical fiber. Conversely, other sensing technologies only give information over the specific points where they are installed, that is to say, they are point sensors. This characteristic of distributed fiber sensors makes them enormously interesting when many points of a structure need to be monitored. In this case, a single distributed fiber sensor can replace many point sensors, which considerably reduces the cost per sensing point when monitoring large structures. In addition, due to the properties of the optical fiber, these sensors have a better performance compared to other kind of sensors. Among other important features, DFOS present a low signal loss, electromagnetic interference immunity, remote sensing and multiplexing capabilities, light weight, and are chemically passive, which make them a very attractive technology for field measurements. Therefore, DFOS have the added advantage of being electrically, magnetically, and chemically passive, so that can be placed in harsh environments, such as nuclear plants or areas with gas concentration, where, due to the possibility of a short circuit, electronic sensors cannot be placed. Al these characteristics make this technology unique. Among the different types of DFOS, those based on stimulated Brillouin scattering, and more specifically, those that build upon the Brillouin optical time-domain analysis (BOTDA) technique, are one of the most promising. The main characteristic that makes BOTDA sensors as promising, is the ability to perform distributed strain and temperature measurements over long distances at high spatial resolution. For the functioning of the sensor, the general interaction that takes place in the BOTDA technique involves two optical waves: a continuous wave probe and a counter-propagating pump pulse. The performance of the sensor response is limited, among others, by the maximum optical power of both waves that can be injected into the fiber. In this way, the main research line in BOTDA sensors is focused on the study of the physical limitations of the technique as well as the development of solutions to these constraints. Another important line relies on the simplification of the sensor setup so as to reduce the complexity and the cost of the sensor. This thesis dissertation contributes to the development of BOTDA sensors by means of different contributions in these two research lines. Several theoretical and experimental studies have been conducted to accurately determine the main limits to the sensor performance in terms of the maximum optical power of the pump and probe waves that can be used. One of the most important limitation in BOTDA sensors is the onset of non-local effects, which limits the maximum pump and probe waves power that can be injected in the fiber, and hence, the signal-to-noise ratio (SNR) at the receiver is worsen. The so-called non-local effects generate measurement errors, because the Brillouin spectra measured at distant locations depend on the interaction at previous positions in the fiber. In this research line, we have examined the effects caused by the limited extinction ratio (ER) of the pump pulse, finding that, among other impairments, it leads to the onset of a new non-local effect originated in the depletion of the pedestal of the pump wave. In addition, it has been found that the pedestal deformation caused by the transient response of erbium-doped fiber amplifiers, which are typically deployed to amplify the pump pulse, also constrains the performance of the sensor. Another contribution is the study of the techniques presented in the literature to mitigate the impairments caused by second-order non-local effects, which cause a frequency-dependent spectral deformation of the pulse. The findings of this study show that these techniques are only applicable when the Brillouin frequency shift (BFS) of the fiber is uniform, which is hard to find in real applications. Lastly, another subject of study is the limitations of the pump and probe optical power in coded-pump wave BOTDA configurations. We have observed that, in addition to some known limitations, there are two important restrictions that have to be taken into account: the onset of non-local effects and the non-linear amplification of the probe wave, both generated by the successive gain induced by the multiple pulses of the coded-pump wave. As a consequence of the findings of these studies, BOTDA configurations intended to solve these limitations have also been proposed during the thesis work. A technique to mitigate the constraints induced by the limited ER of the pump pulse has been presented. This method is based on adding a dithering to the optical source used to generate the two waves involved in the BOTDA sensor, so that the optical wavelength of both signals is modulated. In this way, the Brillouin interaction between the pedestal and the probe wavefronts become uncorrelated, and hence, the influence of the pedestal is greatly reduced. Another contribution is a technique focused on completely overcome the onset of second-order non-local effects. This method is based on continuously tracking the BFS distribution of the fiber, which combined with the probe-dithering method, has allowed, to the best of our knowledge, to inject the highest demonstrated probe wave power in a BOTDA sensor to date. In addition, in order to improve the SNR of the sensor, a novel BOTDA sensor has been proposed. This analyzer combines mono-color cyclic coding and probe-dithering techniques, so that the impairments caused by a coded pump wave are reduced, and hence, it is possible to increase the optical power and consequently enhance the sensing distance range. Finally, a novel simplified BOTDA sensor has been presented, which relies on passive optical filtering of the spectral components generated in a single optical source. In this way, the sensor setup is simplified reducing the number of optical devices, and therefore, the cost of the sensor is also reduced. This BOTDA configuration has been shown to have a performance comparable to more complex setups.