Computation of Viscous Uniform and Shear Flow over A Circular Cylinder by A Finite Element Method

The incompressible viscous uniform and shear flow past a circular cylinder is studied. The two-dimensional Navier-Stokes equations are solved by a finite element method. The governing equations are discretized by a weighted residual method in space. The stable three-step scheme is applied to the momentum equations in the time integration. The numerical model is firstly applied to the computation of the lid-driven cavity flow for its validation. The computed results agree well with the measured data and other numerical results. Then, it is used to simulate the viscous uniform and shear flow over a circular cylinder for Reynolds numbers from lO0 to lO00. The transient time interval before the vortex shedding occurs is shortened considerably by introduction of artificial perturbation. The computed Strouhal number, drag and lift coefficients agree well with the experimental data. The computation shows that the finite element model can be successfully applied to the viscous flow problem.     

A Forced System of Two Cylinders with Various Spacings

The spectrum characteristics and wake structures for a circular cylinder oscillating in a wake are investigated by use of the currently modified virtual boundary method. A forced system of two cylinders with a small spacing ( the downstream one is made to oscillate in the transverse direction) is studied and interesting flow characteristics are observed. A vortex switch and the change of vortex modes (between 2S mode and 2P mode) are observed in the “lock-in“ region. Vortex bands are formed and lost with the increasing excitation frequency. Information concerning saddle points in the flow field is obtained for different excitation frequencies. For a forced system of two cylinders with a large spacing, the upstream cylinder sheds vortexes because there is no downstream cylinder oscillating in the wake. No distinct “lock-in“ response is found for the downstream cylinder.     

Optimization Mathematical Model of Pile Forces for Offshore Piled Breasting Dolphins

An optimization mathematical model of the pile forces for piled breasting dolphins in the open sea under various loading conditions is presented. The optimum layout with the well distributed pile forces and the least number of piles is achieved by the multiplier penalty function method. Several engineering cases have been calculated and compared with the result of the conventional design method. It is shown that the number of piles can be reduced at least by 10%—20% and the piles‘ bearing state is improved greatly.     

An extended variable-grid global ocean circulation model and its preliminary results of the equatorial Pacific circulation

1IntroductionThetropicalPacificOceanplaysanimpor-tantroleintheclimatevariabilitiessuchasElNi-no-SouthernOscillation(ENSO)phenomenon(Chao,1993).ManystudieshavefoundthatthetropicalPacificvariabilitiescanhavesignifi-cantinfluenceontheoceancirculationintheseasadjacenttoChina(Yu,1985;Chaoetal.,1996;Wangetal.,2002).TheseaareaadjacenttoChinaischaracterizedbyitscomplextopog-raphyandnumerousnarrowstraits,andthusre-quiresafinegridtoresolve.Tostudytheinter-actionbetweenthetropicalPacificandChinas…     

Numerical simulation of random wave slamming on structures in the splash zone

The numerical investigation of random wave slamming on superstructures of marine structures in the splash zone is presented in this paper. The impact pressures on the underside of the structure are computed based on the improved volume of fluid method (VOF). The governing equations are Reynolds time-averaged equations and the two equation k– model. The third order upwind difference scheme is applied to the convection term to reduce the effect of numerical viscosity. The numerical wave flume with random wave-maker suitable for VOF is established. Appropriate moving contact-line boundary conditions are introduced to the model wave in contact with and separated from the underside of structure. Parametric studies have been carried out for different incident waves, structure dimensions and structure clearance. The numerical results are verified by the experimental results.     

Modified mild slope equation and open boundary conditions

A numerical model to compute wave field is developed. It is based on the Berkhoff diffraction-refraction equation, in which an energy dissipation term is added, to take into account the breaking and the bottom friction phenomena. The energy dissipation function, by breaking and by bottom friction, is introduced in the Berkhoff equation to obtain a new equation of propagation.The resolution is done with the hybrid finite element method, where lagrangians elements are used.     

Boussinesq-type modelling using an unstructured finite element technique

A model for solving the two-dimensional enhanced Boussinesq equations is presented. The model equations are discretised in space using an unstructured finite element technique. The standard Galerkin method with mixed interpolation is applied. The time discretisation is performed using an explicit three-step Taylor–Galerkin method. The model is extended to the surf and swash zone by inclusion of wave breaking and a moving boundary at the shoreline. Breaking is treated by an existing surface roller model, but a new procedure for the detection of the roller thickness is devised. The model is verified using four test cases and the results are compared with experimental data and results from an existing finite difference Boussinesq model.     

Experimental study of strong reflection of regular water waves over submerged breakwaters in tandem

A series of laboratory experiments was carried out to investigate the strong reflection of regular water waves over a train of submerged breakwaters. Rectangular and trapezoidal shapes of submerged breakwaters are employed and compared for reflecting capability of incident waves. Measured reflection coefficients of regular waves over impermeable submerged breakwaters are verified by comparing with those of the eigenfunction expansion method. A very good agreement is observed. Reflection coefficients of permeable submerged breakwaters are less than those of impermeable breakwaters. The trapezoidal shape is recommended for a submerged breakwater in terms of reflecting capability and practical application.     

A new form of generalized Boussinesq equations for varying water depth

A new set of equations of motion for wave propagation in water with varying depth is derived in this study. The equations expressed by the velocity potentials and the wave surface elevations include first-order non-linearity of waves and have the same dispersion characteristic to the extended Boussinesq equations. Compared to the extended Boussinesq equations, the equations have only two unknown scalars and do not contain spatial derivatives with an order higher than 2. The wave equations are solved by a finite element method. Fourth-order predictor–corrector method is applied in the time integration and a damping layer is applied at the open boundary for absorbing the outgoing waves. The model is applied to several examples of wave propagation in variable water depth. The computational results are compared with experimental data and other numerical results available in literature. The comparison demonstrates that the new form of the equations is capable of calculating wave transformation from relative deep water to shallow water.     

神经网络BP算法在GPS高程转换中粗差探测的尝试

引入神经网络的BP算法中稳定、收敛速度快的Levenberg—Marquardt算法,进行GPS高程转换中的粗差探测,提出了LM粗差探测法。通过实验做出分析比较,得出了实用可行的计算方法。