Journal of Hydrodynamics最新文献

An intelligent flow control on the flow separation over an airfoil under weak turbulent conditions is investigated and solved by deep reinforcement learning (DRL) method. Both single and synthetic jet control at the airfoil angles of attack of 10°, 13°, 15° are compared by training a neural network for closed-loop active flow control strategy based on the soft actor-critic (SAC) algorithm. The training results demonstrate the effectiveness of the deep reinforcement learning-based active flow control method in suppressing the flow separation at high angles of attack, validating its potential in complex flow environments. To improve the stability of the shedding vortex alley over airfoil, a novel reward function considering the vorticity statistics in terms of both vortex and asymmetric shear intensity is first proposed in this work. This vorticity driven reward is demonstrated to perform better in suppressing the rotation and shear intensity and the aerodynamic optimization than the traditional one. Moreover, it can accelerate the convergence speed during the exploration phase. Moreover, it can accelerate the convergence speed during the exploration phase. This study provides valuable insights for future applications of DRL in active flow control under more complex flow conditions.

The KCS Container Ship executing zigzag and turning circle maneuvers in shallow water is studied numerically with CFD simulation. Different factors affecting the accuracy of the numerical prediction such as grid density, overset discretization, propeller model, turbulence modeling and setup condition have been carefully investigated. Numerical predictions for port side and starboard side turning circle maneuvers and 3 different zigzag maneuvers are compared with experimental data obtained by Flanders Hydraulics (FH) for the KCS ship model at scale 1/52.66 with Froude number Fr = 0.095 and water depth to draft ratio h/T = 1.2 in tank conditions. To better assess the accuracy of the numerical prediction, simulations have been performed both with a non-confined and confined configuration.

To illustrate the influence of shear-thinning polymer solution on hydrodynamic and flow structure of element assistance and its mechanism of action, this article conducted a numerical investigation for flow past a cylinder in typical shearing-thinning polymer solutions at Re = 60 based on the finite volume method (FVM). One flexible polymer (PEO) and two rigid polymers (XG, DG), whose rheological properties were experimental fitted using the Carreau-Yasuda model by other literature, were chosen to perform the numerical simulation. The vortex size and the back-flow region length both became smaller in polymer flows, and this inhibition effect was most significant in XG flow. On the contrary, except for the 50 ppm PEO flow, the shedding frequency was promoted in polymer flow. Meanwhile, the three polymers exhibited enhancement effects on the lift coefficient fluctuation, and inhibition effects on the drag coefficient; only XG flow significantly promoted the drag coefficient fluctuation. The dynamic model decomposition (DMD) analysis further indicated that the vortex intensity of each mode in polymer flow was stronger than that in water flow. New structures appeared in PEO, DG flows, while only quantitative difference was found in XG flow compared to water flow. The polymers also affected the mode growth rate and thus the flow stability. The rigid polymer only induced the dispersion degree of the growth rates, while flexible polymer may trigger positive values. In summary, the above hydrodynamic characteristics of shear-thinning polymer flow past a cylinder could provide theoretical support for further understanding the flow characteristics of non-Newtonian fluids.

Underwater structures near the water surface can induce disturbances, leading to wave breaking, which involves complex physical mechanisms. Building on previous studies, this paper conducts a 3-D simulation of breaking waves generated by a NACA0012 hydrofoil at different angles of attack. In contrast to earlier studies, which mostly concentrated on the hydrofoil’s macro-physical parameters like lift and drag, this study focuses on the dynamics of entrained bubbles after free surface breaking. The results indicate that at higher angles of attack, bubbles are swept to greater depths, and both the number and volume of the bubbles increase. By analyzing the bubble velocity data, it is found that the underwater bubble motion is primarily dominated by longitudinal movement, while transverse and vertical bubble velocities are symmetrically distributed. Additionally, vortex structures in the flow field are investigated using the third-generation vortex identification method, Liutex-Omega. It is observed that the vortex structures in the hydrofoil’s wake interact with those generated by free surface breaking downstream, reducing the survival time of large bubbles and increasing the number of small bubbles. Consequently, the bubble number density power law exponent shifts from −10/3–−9/2 as bubble radius increases.

The accumulation of aquatic vegetation in open channels, such as leaves, weeds, and large wood, poses a threat to the safety of hydraulic structures and disrupts the normal operation of hydropower stations, resulting in substantial economic losses. Studying the hydraulic characteristics with vegetation accumulation is a prerequisite for solving this engineering challenge. The effects of vegetation accumulation on the hydraulic characteristics with three types of vegetation were investigated using experimental and numerical simulation methods. The results indicate both backwater rise and head loss increase exponentially with blockage ratio. Furthermore, vegetation accumulation results in an uneven vertical distribution of streamwise velocity, leading to the formation of low-velocity regions and backflow vortexes in the upper water layer. For all cases, the bottom gap velocity increases markedly, forming a high-velocity jet region. The maximum jet velocity remains constant in the near wake region and the decay rate in the far wake region is positively correlated with the blockage ratio. Bed shear stresses in the corresponding region are 1–6 times higher than before vegetation accumulation, increasing the potential for riverbed erosion. This study extends the existing knowledge of vegetation accumulation hydrodynamics to provide a basis for the safe operation of hydraulic structures and river management.

The performance of sewer network is associated with both clean water infiltration and in-sewer pollutant degradation. Quantifying their contributions in large-scale sewer network remains challenging due to the infeasibility of numerous on-site measurements of water flows and water quality concentrations in the whole system. This study developed a physically inverse problem approach to address this challenge, which was tested in an actual sewer network system (25.66 km²) with gridding-based in-sewer flow rate and water quality measurements. Bayesian optimization framework was integrated into sewer hydrodynamic and water quality models to inversely estimate source parameters including source flow rates and source discharge concentrations. Employing simulated annealing algorithm can demonstrate 20.6%–54.2% higher accuracy compared with the other methods, due to its progressive instead of fast and steep convergence toward the true solutions. With the developed approach, the infiltrated clean water infiltration and mass loss of chemical oxygen demand (COD) within the sewer network were quantified synchronously. Further, the condition of sewer structural defects was assessed, and a reference value for allowable in-sewer COD degradation was also presented, which was 4%–5% COD mass per hour of sewage hydraulic retention. Therefore, this methodology can provide cost-effective solution for comprehensive assessment of sewer network conditions.

In this paper, a multi-objective optimization strategy for a volute mixed flow pump is implemented based on inverse design theory. The controllable velocity moment (CVM) design parameters are defined based on the impeller outlet velocity moment and its partial derivative in the streamline direction, and flow control is further realized by directly adjusting the velocity moment distribution inside the impeller. The significance order of the effect of the CVM parameters on the efficiency and NPHS_c at the best operation points is first investigated, and the elliptical basis function (EBF) approximation model and MOPS optimization algorithm are combined to carry out multi-objective optimization. Compared with the baseline model, an improvement in the efficiency and head at the best operation points of 0.48%, 1.07 m is obtained for the optimal model, with a widened and efficient operation range, and NPHS_c is reduced from 4.545 m–4.235 m, with a slowed cavitation development. The proposed CVM method effectively realizes flow control under multiple conditions, optimizing the pressure distribution on the blade pressure side and suppressing the reverse jet in the cavity closure area.

In the present paper, the simultaneous resonance of a cylindrical bubble under triple-frequency acoustic excitation is investigated theoretically. Specifically, based on the multi-scale method, the dimensionless oscillation equations and the second-order analytical solutions of the primary-subharmonic-subharmonic (PRI-SUB-SUB), primary-superharmonic-superharmonic (PRI-SUPER-SUPER) simultaneous resonances are obtained. Based on the analysis of the frequency response, the nonlinear dynamic behavior of the cylindrical bubble and its influencing factors are analyzed. The primary conclusions include: (1) Under triple-frequency acoustic excitation, the frequency response of PRI-SUB-SUB presents a single peak, and that of PRI-SUPER-SUPER presents two peaks. (2) The polytropic exponent affects both the peak value and position of the resonance peak in the frequency response. (3) The unstable region in frequency response curve of the simultaneous resonance is significantly affected by the total amplitude and equilibrium radius, presenting a positive correlation.

The misoperation of hydraulic components such as pumps and valves in pressurized pipelines triggers water hammer phenomena and seriously threats the safe operation of hydraulic systems. At present, the main water hammer simulation methods are method of characteristics (MOC), and further investigation of new algorithms is needed. Therefore, a new method for simulating the water hammer using the finite volume method (FVM), semi-implicit method for pressure linked equations (SIMPLE) algorithm is proposed in the present work. Compared with the experimental data, the accuracy and reliability of the proposed algorithm are verified. Results show that the IAB, MIAB friction models not only predict the first pressure peak but also accurately predict the pressure attenuation. From the comparison of the MOC, SIMPLE algorithms, the results of the two algorithms are almost the same in front of the valve, while near the upstream tank, when using the same friction model, the pressure attenuation predicted by the SIMPLE algorithm is slightly greater than that of the MOC method and closer to the experimental data. Therefore, the newly proposed algorithm can serve as an alternative to the MOC method in simulating water hammer. The investigation enriches the numerical methods of hydraulic transients and lays the foundation for subsequent program development.

The cavitation bubble precipitation refers to the formation process of the spherical cavities, known as cavitation bubbles, as the ambient pressure of water decreases. In the fields of hydraulic machinery, the saturated vapor pressure of clean water is often used as the pressure threshold for cavitation occurrence. However, the engineering practice has demonstrated that, the incipient cavitation pressure may be significantly higher than the saturated vapor pressure, especially in sand-laden water conditions. Therefore, to determine a reasonable cavitation pressure threshold and ensure the accurate cavitation flow simulations and effective assessment of cavitation risks for hydraulic machinery operating in sand-laden water conditions, an experimental investigation is conducted. First, a high-precision experimental setup based on the vacuum pump, high-frequency pressure sensor and high-speed camera is constructed. This setup allows for the continuous pressure reduction in water, acquisition of high-precision pressure data and tracking of the entire cavitation bubble precipitation process. Second, based on the experiments in clean water conditions, the relationship between the cavitation bubble precipitation degree and pressure is established, and two key states of incipient cavitation and boiling cavitation are defined. Third, based on the experiments in sand-laden water conditions, it is observed that the numerous cavitation nuclei on sand surfaces make both the incipient and boiling cavitation pressure in sand-laden water higher than those in clean water. The quantitative relationship between the sand concentration and diameter, and the cavitation pressure is established, providing a more reasonable cavitation pressure threshold. This investigation enhances the understanding of cavitation bubble precipitation in sand-laden water and supports the development of more accurate cavitation models for hydraulic machinery operating in sand-laden water conditions.