No abstract

No abstract

Experiments have been conducted to investigate the two-degree-of-freedom vortex-induced vibration (VIV) response of a rigid section of a curved circular cylinder with low mass-damping ratio. Two curved configurations, a concave and a convex, were tested regarding the direction of the flow, in addition to a straight cylinder that served as reference. Amplitude and frequency responses are presented versus reduced velocity for a Reynolds number range between 750 and 15 000. Results for the curved cylinders with concave and convex configurations revealed significantly lower vibration amplitudes when compared to the typical VIV response of a straight cylinder. However, the concave cylinder showed relatively higher amplitudes than the convex cylinder which were sustained beyond the typical synchronisation region. We believe this distinct behaviour between the convex and the concave configurations is related to the wake interference taking place in the lower half of the curvature due to perturbations generated in the horizontal section when it is positioned upstream. Particle-image velocimetry (PIV) measurements of the separated flow along the cylinder highlight the effect of curvature on vortex formation and excitation revealing a complex fluid–structure interaction mechanism

No abstract

This paper presents a study of single‐channel proton exchange membrane fuel cells (PEMFCs) using computational modeling and simulation. For this analysis, the commercial software COMSOL Multiphysics was used to build a single‐phase isothermal and tridimensional fuel cell model. For the mathematical description of the catalyst layer, the flooded agglomerate model was implemented, and it proved to be more accurate than Butler‐Volmer model, which is the pre‐built model in the software. Such evidence was verified when comparing the polarization curves obtained using both models with an experimental curve. After definition of the model, the main objective of this study was to analyze the influence of the flow channel cross‐section in the water distribution inside the cell, studying rectangular, trapezoidal and hybrid stepped geometries. The fuel cell with stepped channel was equivalent to the trapezoidal cell in all aspects analyzed, and both provided superior water management than the rectangular cell. However, the current generation in the rectangular design was slightly higher. It was noted that the simulation of a tridimensional model provided a better understanding of the regions where higher concentrations of water can occur, and that different flow channel designs can be used to enhance water management.

This paper presents the modeling, simulation and optimization of a single channel proton exchange membrane fuel cell (PEMFC) using computational fluid dynamics methods. The shape optimization of the flow‐channels was employed to improve the electrical performance of the fuel cell. The maximization of the current density was the objective function of the single‐objective optimization problem, the upper and lower widths of the flow channels were the control variables and a cross‐section area restriction was imposed. The optimized flow‐channel PEMFC presented improved current generation characteristics, showing higher current and power densities and a more uniform current density distribution than the rectangular flow‐channel.

This paper presents experimental results concerning the response of circular cylinders with and without strakes. The longitudinal and transverse fluid forces (drag and lift), amplitude response and wake structures of plain and helically straked cylinders are compared. Six different configurations of straked cylinders with pitches (p) equal to 5D, 10D and 15D and heights (h) equal to 0.1D and 0.2D are investigated. Measurements on the dynamic response oscillations of an isolated plain and straked cylinders and flow visualization employing a PIV system are shown. Fixed cylinder drag measurements are also shown. The models are mounted on an elastic base fitted with flexor blades and instrumented with strain gauges or in an air bearing base. The base is fixed on the test-section of a water channel facility. The flexor blades possess a low-damping and the flexor blades base an the air bearing base are free to oscillate only in the cross-flow direction. The Reynolds number of the experiments ranges from 2000 to 10000, and reduced velocities, based on natural frequency in still water, vary up to 13. The drag coefficient is increased by 20% for the h = 0.1D cylinder, and 60% for the h = 0.2D cylinder, comparing both with the plain cylinder. The smaller height strokes (h = 0.1D) do not prevent vortex formation in the region very close to the body, resulting in a decrease of about 50% of the amplitude response compared with the plain cylinder. Lowest amplitude response was found to the p = 10D and h = 0.2D case. The analysis of the vorticity contours shows that the shear layer does not roll close to the body (same result for the other cases with h = 0.2D).

Two- and three-dimensional simulations of the flow around straked cylinders are presented. For the two-dimensional simulations we used the Spectral/hp Element Method, and carried out simulations for five different angles of rotation of the cylinder with respect to the free stream. Fixed and elastically-mounted cylinders were tested, and the Reynolds number was kept constant and equal to 150. The results were compared to those obtained from the simulation of the flow around a bare cylinder under the same conditions. We observed that the two-dimensional strakes are not effective in suppressing the vibration of the cylinders, but also noticed that the responses were completely different even with a slight change in the angle of rotation of the body. The three-dimensional results showed that there are two mechanisms of suppression: the main one is the decrease in the vortex shedding correlation along the span, whilst a secondary one is the vortex wake formation farther downstream.