Assessment of Fiber Bragg Grating Sensors for Monitoring Shaft Vibrations of Hydraulic Turbines

Abstract

The structural dynamic response of hydraulic turbines needs to be continuously monitored to predict incipient failures and avoid catastrophic breakdowns. Current methods based on traditional off-board vibration sensors mounted on fixed components do not permit inferring loads induced on rotating parts with enough accuracy. Therefore, the present paper assesses the performance of fiber Bragg grating sensors to measure the vibrations induced on a rotating shaft–disc assembly partially submerged in water resembling a hydraulic turbine rotor. An innovative mounting procedure for installing the sensors is developed and tested, which consists of machining a thin groove along a shaft line to embed a fiber-optic array that can pass through the bearings. At the top of the shaft, a rotary joint is used to extract, in real time, the signals to the interrogator. The shaft strain distribution is measured with high spatial resolution at different rotating speeds in air and water. From this, the natural frequencies, damping ratios, and their associated mode shapes are quantified at different operating conditions. Additionally, the change induced in the modes of vibration by the rotation effects is well captured. All in all, these results validate the suitability of this new fiber-optic technology for such applications and its overall better performance in terms of sensitivity and spatial resolution relative to traditional equipment. The next steps will consist of testing this new sensing technology in actual full-scale hydraulic turbines.

An Improved Spectral Subtraction Method for Eliminating Additive Noise in Condition Monitoring System Using Fiber Bragg Grating Sensors

“…Compared with the OFDR and OTDR, the FBG sensor has the advantages of high sensitivity to high-frequency vibrations, high accuracy for monitoring vibrations at fixed points, and cost-efficient interrogators [ 11 ]. Thus, the FBG sensor is more suitable for monitoring crucial components of rotating machinery and has been widely applied to vibration measurement of rotating machinery [ 12 , 13 , 14 ]. However, the weak fault component in the complex measured signal may not be accurately distinguished because the measured signal is corrupted by a large amount of additive noise in the CM system using FBG sensors, which indicates that it is difficult to extract the useful and meaningful fault information.…”