Hydraulic stability is one of the key problems during the design stage of hydraulic turbines. Despite of modern computational tools that help to define dangerous operating conditions and optimize runner design, hydraulic instabilities may fortuitously arise during the turbine life, as a consequence of variable and different operating conditions at which a hydraulic turbine can be subject.
In general, the presence of unsteady flow reveals itself in two different ways: at small flow rate, the swirling flow in the draft tube conical inlet occupies a large portion of the inlet and causes a strong helical vortex rope; at large flow rate conditions the unsteady flow starts midway and causes a breakdownlike vortex bubble, followed by weak helical waves.
In any case, hydraulic instability causes mechanical effects on the runner, on the whole turbine and on the draft tube, which may eventually produce severe damages on the turbine unit and whose most evident symptoms are vibrations. This notwithstanding, condition monitoring systems seldom are installed on this purpose in hydraulic power plants and no examples are reported in literature about the use of model-based methods to detect hydraulic instability onset.
In this paper, by taking the advantage of a testing campaign performed during the commissioning of a 23 MW Kaplan hydraulic turbine unit, a rotordynamic model-based method is proposed. The turbine was equipped by proximity and vibration velocity probes, that allowed measuring lateral and axial vibrations of the shaft-line, under many different operating conditions, including also some off-design ones. The turbine mechanical model, realized by means of finite beam elements and considering lateral and axial degrees of freedom, is used to predict turbine unit response to the unsteady flow. Mechanical system response is then compared to the measured one and the possibility to detect instability onset, especially in real-time, is discussed.
Correction: Characterisation of Mixing in the Proximal Duodenum of the Rat during Longitudinal Contractions and Comparison with a Fluid Mechanical Model Based on Spatiotemporal Motility Data
