Search

The clearance flow patterns and pressure distribution determine the clearance axial hydraulic thrust of a pump turbine, which has a substantial impact on the unit axial imbalance. However, due to the tiny size and complex shape of the clearance flow channel, there is no detailed description of the flow patterns and pressure characteristics. In this study, we conducted a model test with particle image velocimetry (PIV) measurements and CFD simulation of a rotating disk flow that was a simplified model of the pump-turbine clearance flow. It is shown that a typical Batchelor flow is formed in the clearance region, demonstrating a “core region + double-boundary layers” distribution for the circumferential velocity along the clearance height direction; the core region rotates at a speed of only around 41–42% of the rotating disk speed and is independent of the clearance inlet pressure and clearance height. Driven by centrifugal force, the flow is radially outward around the rotating disk, while inward around the stationary disk in the meridian section, showing secondary flow vortices. The pressure in the clearance region has a circumferentially symmetric and radially quadratic distribution. Based on the liquid differential equilibrium equation and core region circumferential velocity, the pressure and clearance axial hydraulic thrust in the clearance region can be expressed as a function of the clearance inlet pressure and the square of the runner rotating speed. These findings can be used to investigate axial force imbalance issues of the pump-turbine unit.

To meet the demand of the power grid for flexible regulation, pump turbines have to frequently and fleetly stop, start, and transition between different modes, causing many instability problems. To solve the problems, understanding the complex flow and pressure pulsation features in different working conditions is the prerequisite. In this study, the flow patterns, pressure pulsations, cavitation characteristics, and runner forces of a prototype pump-turbine working under the five operating modes were displayed and analyzed based on computational fluid dynamics simulations. It is shown that when working points are near the pump and turbine design working points, the flow is smooth, with small pressure pulsation and without cavitation; however, when working points deviate from the optimal ones, flow patterns become worse and cavitation appears. In the pump-brake mode, the flow hits the runner blades violently, causing obvious pressure and runner force fluctuations. In the hump region, the flow has a strong impact on the guide vanes, and the rotating stall vortices are apparent in the vane area. In the S-shaped region, pressure pulsations are the highest with prominent rotor–stator interaction and rotating stall. Under similar working conditions, a larger guide-vane opening corresponds to a larger pressure pulsation amplitude. On the spatial distribution, the largest pressure pulsations happen in the vaneless space.

LiAlSiO4 (LAS) ceramics are prepared by using the sol-gel method followed by spark plasma sintering. XRD patterns and SEM images verify that the ceramics contain amorphous and LAS phases and that microcracks appear in the sample prepared at 900 °C due to its larger grain size. Compared with applied pressure and soaking time, sintering temperature has a greater impact on the crystallinity and density of the ceramics during sintering. High-temperature XRD results reveal that the LAS phase exhibits its intrinsic negative thermal expansion independently in all samples regardless of crystallinity. The coefficients of thermal expansion (CTE) measured by the dilatometric method change from positive values in samples prepared at 600 and 650 °C to near zero in samples prepared at 700 and 800 °C and then to a negative value in the sample prepared at 900 °C. The combined effects of an amorphous phase with a positive CTE and the LAS phase with a negative CTE are responsible for the observed transformation of thermal expansion in the samples. The calculated total CTEs of the glass-ceramic bulks are in agreement with the results measured through the dilatometric method in samples prepared at 650–800 °C. Microcracks in the sample prepared at 900 °C cause a more negative bulk CTE than the calculated CTE.

Monoclinic K0.55WO3.275 has been synthesized by solid state reaction method. The space group is P2/m with cell parameters a = 15.9753(2) A, b = 7.73149(8) A, c = 10.9393(1) A and β = 108.477(1) ° at 313 K. The intrinsic linear coefficients of thermal expansion calculated by temperature-dependent X-ray diffraction data are 14.70 × 10−6 K−1 (253–323 K), 3.38 × 10−6 K−1 (323–363 K) and 11.52 × 10−6 K−1 (363–433 K), respectively, which are consistent with the apparent ones obtained by thermal dilatometer. The a axis shows positive thermal expansion from 253–433 K. The b and c axes display negative thermal expansion in the range of 323–363 K, and positive thermal expansion in two ranges of 253–323 K and 363–433 K. Three W–O–W bending modes with negative Gruneisen parameter may be responsible for the low thermal expansion in the range of 323–363 K.