岛礁地形上二维波浪传播的数值模拟

波浪在岛礁地形上传播的重要特征之一是波浪的破碎,为了研究近岸岛礁地形下波浪传播特性,本文采用基于Boussinesq方程的FUNWAVE1.0模型,对于二维岛礁地形上波浪的传播过程进行了数值计算,通过数值计算结果与物模实验结果的对比分析,研究了数值计算模型中波浪破碎模拟参数的取值,研究结果可为实际工程波浪破碎过程的模拟分析提供参考。     

用一种全非线性Boussinesq方程模拟斜坡上的波浪破碎和爬高

基于一种新的二阶全非线性Boussinesq方程,采用预测-校正格式的有限差分法对该方程进行离散,建立了数值模型.模型中通过“狭槽法“来处理波浪在岸线处的动边界条件,采用涡粘模型来模拟波浪破碎引起的能量耗散.为了验证数值模型的适用性,模拟了斜坡地形上的波浪破碎和爬高.从数值结果和试验结果的比较上看,该模型可以很好地模拟近岸波浪场的实际问题.     

非交错网格下Boussinesq方程的源函数产生波浪的方法

针对由Beji型的Boussinesq方程推导出的在计算域内产生波浪的源函数,应用有限差分法的Adams-Bashforth-Moulton预报校正格式,在非交错网格下进行数值模拟。在一维方向上进行离散,对一维平底地形和潜堤地形进行了数值模拟,数值计算结果与物理模型试验结果吻合较好,表明采用源函数的方法可以很好地在数值计算域内产生波浪。     

波浪作用下单体网格式浮架受力的数值模拟

模型试验表明重力式网箱在波浪作用下，单纯浮架的受力贡献比例较大。本文运用波浪理论，通过建立浮架、锚绳及浮子的运动方程，采用五阶龙格-库塔进行数值求解，得到了单体网格式锚碇单纯浮架的数值计算模式。将计算模型简化为四点锚碇，并将计算结果与实验结果进行了比较，二者符合程度较好。在此基础上，比较了单体网格式锚碇与四点锚碇单纯浮架的受力，结果表明网格式锚碇方式浮架受力较小。     

BW模型在掩护水域内波浪计算中的应用

采用MIKE21-BW模型分别对平底地形下单突堤的波浪绕射以及实际地形下某港港内波浪场进行模拟及验证,将模拟结果与规范中的计算图解及物模试验结果进行对比分析,吻合良好。同时研究发现,由于该模型数值格式的限制,在波浪斜向入射的情况下,模型的计算结果存在较为明显的误差,而经过地形逆向旋转处理,将波浪入射方向调整为正向入射的方法可以较好地改善模型的计算结果。另外,在某港实际地形下的波浪模拟中,还探讨了谱峰因子γ的变化对该港港内波浪场的影响。     

港湾内非线性波浪对船剖面作用的时域计算模型(英文)

文中用边界元数值求解复速度势模拟了波浪对船剖面的时域作用力,采用满足运动学和动力学条件的半拉朗格朗日法跟踪流体的自由表面.推导了适用本数值模型的辐射边界条件,此边界条件和海绵层边界条件组成的出口边界条件,可以使波浪很好地被吸收或传播出去.进行了相关物理模型实验,数值计算结果和实验结果的对比表明此模型能精确模拟波浪对船的作用力.     

基于数值和实验方法的非透浪梳式防波堤的水动力学特性研究

采用源函数造波法建立了三维数值波浪水槽模型,模拟了不同随机种子数(NW)下的随机波浪,与目标谱对比的结果证明当NW=200时,采用文中的数值方法可以得到较好的模拟精度;建立了随机波浪对一种非透浪梳式防波堤作用的数值模型,通过数值模拟结果和实验结果的比较,验证了该数值模型的有效性。对该非透浪的梳式防波堤的水力学特性进行了实验研究,并应用上述数值方法对结构的所受冲击波浪力机理进行了分析,数值结果证明在该结构的危险水位下,由结构的翼板和胸墙下底板所构成的异型空腔结构是导致翼板上产生较大冲击压力的主要因素。在此基础上,为了消减翼板的冲击压力,提出一种改进的结构型式,最后对该改进结构的翼板上波浪力特性和波浪反射系数进行了实验研究。     

开孔率对开孔板消浪效果影响的数值模拟研究

采用VOF方法,结合k-ε紊流模型建立数值波浪水槽。造波边界采用内源造波法,造波边界后方和波浪水槽出口边界均采用海绵层消波,自由面采用F函数追踪,从而数值模拟了波浪对开孔板的作用。将开孔板前波浪反射系数的数值计算结果与文献[14]的物理模型实验结果进行比较,验证所建立的数值模型及计算方法的正确性。通过改变开孔率,模拟波浪与开孔板作用的开孔板前波浪的反射系数,进而分析了开孔率与波浪反射率的关系。同时,分析了波浪作用下开孔板的迎浪面与背浪面的点压力差变化,并主要探讨开孔率与点压力差的相互影响关系。分析结果表明:随着开孔率α的增大,反射率减小,透射率增大,点压差呈非线性变化。     

基于OpenFOAM的船舶与液舱流体晃荡在波浪中时域耦合运动的数值模拟

波浪中载液船舶运动激励舱内液体的晃荡,舱内液体晃荡产生的冲击力同时作用在舱壁上,进而影响船舶的运动姿态。波浪中船体水动力和时延函数是在势流理论范畴下采用切片法和脉冲响应函数方法计算获得的,液舱内液体非线性晃荡是基于粘性流理论实时计算模拟,两者耦合建立了波浪中载液船舶与液舱流体晃荡耦合的运动方程。论文基于开源CFD开发平台OpenFOAM,自主开发实现了船体运动与液舱晃荡的耦合计算程序,并进行了相应的数值模拟计算和验证工作。该方法完整地考虑了波浪、船体和液舱晃荡之间的耦合作用,并结合船体内外流场特点分别采用了势流和粘性流理论,具有较高的计算效率。通过数值模拟计算和模型实验研究表明,数值模拟计算能够清晰显现出液舱晃荡对船体全局运动影响,船体运动计算结果与模型实验结果吻合良好。     

基于抛物型缓坡方程的波浪数值模型应用研究

在建立了基于抛物型缓坡方程的波浪变形数值预报模型的基础上,计算了波浪破碎前堤后绕射及浅滩上的折射现象。在计算堤后绕射中,采用了堤上薄膜水的方法。为了消除数值计算结果的高频振荡,在方程中引入耗散界面。浅滩折射的数值结果还与Berkhoff的经典试验结果进行了对比。讨论了摩阻系数对波高的影响。为了将模型运用于破碎区,提出了一种波能衰减模型,分别用Sato的水槽试验资料及实际海域破波区波高实测资料对模型进行了验证。上述计算结果表明,所建立的波浪变形数值模型可以用于近岸波浪变形计算中。     

交错网格下四阶Boussinesq方程对潜堤上波浪演化的应用*

基于Madsen 和 Sch?ffer推导的四阶Boussinesq方程,在交错网格下建立一维数值模型。模型中的时间导数采用混合四阶Adams-Bashforth-Moulton 格式。为了避免二次反射,采用了Gobbi 和 Kirby的的内部源函数造波法。最后数值模拟了波浪在潜堤上的演化过程,并且与Luth的试验值进行对比,结果显示两者吻合较好,另外还给出了非交错网格下的模拟结果,通过对比可知,两个模型都能够很好地模拟波浪在潜堤上的演化过程。     

Application of large eddy simulation in the performance study of wave blocker

The configuration and aerodynamic performance of the inlet system are important aspects in the process of installing a gas turbine on a naval vessel. Under the requirements, large eddy simulation （LES） is used to simulate the three-dimensional fluid flow in the wave blocker of a marine inlet filter. The Smagorinsky-Lilly sub-grid model was used to model motions of small-scale structures. During numerical simulation, the SIMPLE algorithm was applied. The central-differencing spatial discretization scheme and the second order accuracy finite difference for the temporal discretization were used. Simulation gives satisfactory distribution of the vorticity fields and turbulent kinetic energy. Compared with the k-ε turbulent model, the results of LES are better for the distribution of parameters. The results of experimental study in a small-scale wind tunnel indicate that numerical calculation has higher accuracy. Therefore, the methods used are worthy of reference and introduction for the design of an inlet system.     

基于细长体公式的Truss SPAR在波流中的运动响应预报方法

利用细长体公式和刚体非线性运动方程建立Truss SPAR在波浪与流中运动响应预报方法。通过波浪自由表面和SPAR中心线方程构造辅助函数,迭代计算Truss SPAR瞬时湿长度。根据SPAR主体形状特点和流场中水质点运动规律,分段高效积分Truss SPAR上的水动力载荷。通过Runge-Kutta-Fehlberg方法求解运动方程,得Truss SPAR在波浪与流中的运动响应。对一座Truss SPAR在不同波流工况中的运动响应进行了预报,结果显示波浪和流使Truss SPAR产生了明显漂移运动和振荡运动,漂移运动的大小与流的方向有关,而振荡运动的幅值与波浪的方向有关。     

A finite pointset method for the numerical simulation of free surface flow around a ship

In this paper, we present a Finite pointset method (FPM) for the numerical simulation of free surface flow around a ship in calm water. It is a Lagrangian and meshless particle scheme which is applied to the projection method for the incompressible governing equations. This requires the solution of Poisson problems in each time step, so a moving least squares (MLS) interpolants is used for the spatial derivatives in order to discretize the Poisson equation with pressure-Dirichlet condition of free surface flow in meshless structure. Meanwhile, an additional problem of the periodic particle locations redistribution in the present approach is still handled by MLS interpolants. With the proposed FPM technique, problems associated with the free surface flow around a ship are circumvented. A verification of numerical modeling is made using the Wigley hull and the validity of the proposed methodology is examined by comparing the detail of wave profile and wave-making resistance with Series 60 model. The results demonstrate that FPM is able to perform efficient and stable simulations of free surface flow around a ship.     

Numerical simulation of violent sloshing by a CIP-based method

In this article, a new computational fluid dynamics simulation approach based on the constraint interpolation profile (CIP) method is applied to tackle the violent sloshing problem. The present study considers two-dimensional sloshing phenomena in a rectangular tank. By the proposed method, the sloshing problem is viewed as a multiphase problem that includes water and air flows. A stationary Cartesian grid is used and the free surface is solved by an interface capturing method. The CIP combined unified procedure (CCUP) scheme was adopted for the flow solver, and both the CIP scheme and the CIP conservative semi-Lagrangian with cubic interpolation polynomial (CIP-CSL3) scheme were used for interface capturing. For validation of the numerical method, a physical experiment was conducted with a rectangular tank for several frequencies and filling heights. A convergence check was first performed for the method. The numerical simulation results on violent sloshing show that the use of the CIP-CSL3 scheme as an interface capturing procedure gives much better results for the pressures and free-surface profiles than the conventional CIP scheme.     

一种处理高阶面元法中奇异积分的方法

对高阶面元法中奇异积分问题进行数值研究,根据面元的大小以及面元到场点的距离,把整个曲面积分分为远场、近场两类,分别对其使用不同的方法以处理Rankine源项所引起的积分奇异问题,对三维球体、椭球体进行数值计算,将结果与解析解和其他方法的计算值的比较表明此计算方法是有效的。     

基于双层Boussinesq方程的聚焦波数值模型

基于高精度双层Boussinesq方程,建立了聚焦波的时域波浪数值水槽。时间积分采用混合4阶Adams-BashforthMoulton预报-校正格式,聚焦波生成则采用累加不同频率规则波的内部造波源项法。针对Baldock等的聚焦波试验进行数值计算,计算结果与试验数据吻合较好。利用验证后模型进一步考察了非线性对数值计算聚焦波的影响,其中考虑了强非线性、弱非线性以及线性3种情况,结果表明非线性对精确模拟聚焦波至关重要,强非线性模型给出的结果最好,弱非线性次之,线性最差。     

Advection schemes for shelf sea models

Numerical models of shelf seas must handle sharp gradients in thermoclines and fronts, and at the edges of patches of passive tracer, often in the presence of strong tidal currents, without introducing excessive numerical diffusion or spurious oscillations. In a sigma coordinate model there is an additional problem, since then purely horizontal motion over a sloping sea bed may introduce strong numerical diffusion in the vertical, which can artificially erode thermoclines.In this paper, two advection schemes applicable to shelf sea models are examined. They are based on TVD (Total Variation Diminishing) and PPM (Piecewise Parabolic Method) techniques. They are demonstrated to give satisfactory performance for tracer advection in one, two and three dimensions. They are both monotonic, and the PPM scheme has particularly low numerical diffusion. In two and three dimensions directional splitting is used, which is handled by following advection with each horizontal velocity component by adjustment of the sigma levels in each water column using the same advection scheme. The artificial vertical diffusion is then small, particularly with PPM.The two schemes are also compared in a three-dimensional model of a cylindrical eddy of relatively fresh water, released from rest in an open sea region. Here, both salinity and momentum are advected. Laboratory experiments show that after an initial period of adjustment the eddy should become unstable, with a growth of cyclonic-anticyclonic eddy pairs. This is reproduced in the model, with the PPM scheme again producing sharper results than the TVD scheme.Integrated second moment calculations are used to compare the schemes. These demonstrate the lower numerical diffusion of the PPM scheme. This advantage is achieved at the cost of greater computing time.     

DTMB5415驱逐舰绕流的数值模拟

本文采用RANS方程及RNG湍流模型数值模拟了带方尾及声纳罩的5415驱逐舰绕流，计算结果表明对复杂的实际船形，CFD能有效地预报船体周围流场、波形和阻力等船舶设计感兴趣的水动力性能参数，为船形分析及优化提供了有力的工具。