We present a demodulation method that partially compensates for the nonlinear and random transfer function of fiber-optic direct detection coherent Rayleigh optical time-domain reflectometry (C-OTDR). Specifically, the proposed method is shown to improve the detection of small amplitude, high frequency dynamic optical fiber strain as it occurs in acoustic emission sensing. The method is applicable provided that the dynamic fiber strain to be sensed is known to affect a longer section of sensing fiber in a spatially homogeneous way. It is shown that this knowledge can be used to extract more quantitative information from the measured C-OTDR signal by averaging signal components from the affected fiber section in a suitable and efficient way. The theoretical basis of the method is developed and supporting experimental results are presented.
We propose the application of Distributed Acoustic Sensing (DAS) based on Rayleigh Coherent Optical Time-Domain Reflectometry (C-OTDR) to unconventional sensing tasks in industrial condition monitoring. As examples we present results on the way to fibre-optic remote sensing of dielectric damage processes in high voltage cable joints as well as to condition monitoring of passive rollers in large industrial belt conveyor systems.
We report on the successful application of distributed acoustic sensing (DAS) to the detection of partial discharge (PD). A detection limit of about 1 nC discharge magnitude was achieved for PD in a real-scale model of a high voltage termination. Dedicated ultrasonic fibre-optic transducers were interrogated using coherent optical time-domain Rayleigh backscatter reflectometry (C-OTDR). Random quadrature demodulation was employed for retrieving relevant acoustic information from the raw C-OTDR backscatter traces. To our knowledge, our results are a first-time demonstration that quasi-distributed fibre-optic acoustic sensing is a candidate technology for the acoustic partial discharge monitoring of power cable joints and terminations.
We have shown that for C-ODTR based DAS two UV-inscribed scattering segments within an optical fiber have a significant influence on the local sensitivity Fading characteristics. This method was experimental proven to be suitable for distributed dynamic strain measurement in a thermally unstable environment. Signal-to-noise ratio analyses for detecting a signal peak within the time-dependent spectra were performed. Results show an increase in SNR and additionally that the occurrence of total loss of sensitivity (SNR <1) could be decreased by a factor of six when compared to an untreated standard fiber. We have shown that for C-ODTR based DAS two UV-inscribed scattering segments within an optical fiber have a significant influence on the local sensitivity Fading characteristics. This method was experimental proven to be suitable for distributed dynamic strain measurement in a thermally unstable environment. Signal-to-noise ratio analyses for detecting a signal peak within the time-dependent spectra were performed. Results show an increase in SNR and additionally that the occurrence of total loss of sensitivity (SNR <1) could be decreased by a factor of six when compared to an untreated standard fiber.