Pressure Coefficient Distribution Research Articles

Kinetic description of the turbulence in the supersonic compressible flow over a backward/forward-facing step

In the present paper the Boltzmann kinetic equation is applied to investigate a supersonic turbulent compressible flow. For this purpose a numerical simulation of supersonic compressible two-dimensional flow in the vicinity of a backward/forward-facing step with leeward-face angle ranging from 8° to 90° and free-stream Mach number Ma∞≈3 is carried out. An explicit–implicit numerical scheme in the framework of the direct approach for the solution of the full Boltzmann equation is used. Principal features of the kinetic approach for the description of compressible turbulence are discussed. The comparison with experimental and numerical results presented in the open literature is performed in terms of pressure, skin friction and heat-transfer coefficient distributions over the step surface. Kinetic computed flow field structures for different leeward-face angles are presented and discussed. A good agreement with the experimental results is observed.

Modelling the Flow in a Transonic Diffuser with One Reynolds-Stress and Two Eddy-Viscosity Models

Turbulence modeling combined with the mass averaged Navier-Stokes equations and detailed comparisons with experimental data are presented for a transonic flow in a converging diverging diffuser for two different inlet Mach numbers. Three low-Reynolds number turbulence models, which do not use the concept of the wall distance and wall normal vector calculation are adopted: a linear eddy-viscosity model, a cubic non-linear eddy-viscosity model and a non-linear Reynolds-stress model. The compressibility effects due to the mean density variations are incorporated in a subsonic pressure-based flow solver through minor modifications to the pressure velocity-coupling algorithm and by introducing the dilatation-dissipation of turbulence to the equations of the turbulent models. Grid dependency studies were conducted and detailed velocity and pressure coefficient distributions are presented in comparison with available experimental data providing also information for the boundary layer development on the diffuser walls in the region downstream the shock wave. The final results are very promising concerning the capability of the turbulence models to capture the compressibility effects in transonic flows when they are incorporated with appropriate modifications in a pressure-based flow solver.

二次元オフセット噴流の流動特性と熱伝達

The flow field of an offset jet is complex and is frequently found in many engineering devices. In this paper, the fluid flow and heat transfer characteristics of two-dimensional offset jet are investigated experimentally. The offset jet is produced by the flow of air which issue from the end of a long parallel channel. The exit Reynolds number Re is changed in 1.5×103 to 7.0×103, and the offset ratio H/h (H : step height, h : channel height) is changed in 0.5, 1.0 and 1.5. The wall shear stress is measured using a micro flow sensor. Flow visualization is carried out using a CCD camera and laser light. The velocity profiles are measured by a PIV system. When the flow at the channel exit is laminar, fluid in the outer region has a counter-clockwise vorticity and fluid in the inner region has a clockwise vorticity, and a lifting-off of the wall jet flow into the ambient fluid is occurred. The Strouhal number of the vortex frequency is almost constant against the offset ratio. The formation of those vortices is not seen at the turbulent flow. Two peaks exist in the wall static pressure, Nusselt number and skin friction coefficient distributions of the laminar flow with H/h = 0.5 and 1.0. The 1st peak appears near the reattachment point, and the 2nd peak appears in the neighborhood of the lifting-off of the wall jet flow. The two peaks coalesce into one at H/h = 1.5. As the offset ratio increases, the reattachment point moves to downstream direction and Nu becomes large both the laminar and turbulent flows. The reattachment point of the turbulent flow moves to upstream direction compared with the laminar flow because Coanda-effect is strengthen.

低雷诺数旋翼翼型设计及气动仿真

This research focuses on the thin circular arc rotor airfoils at a low Reynolds number.A novel thin circular arc airfoil with a convex structure was designed in consideration of its demand for high aerodynamic performance,high structural strength,lightweighting and easy manufacture.A convex curve on the upper surface of the airfoil was adopted to increase the thickness of airfoil at the partial chord and a stiffener in the airfoil was installed to improve the structural strength of blade span wise.The designed thin circular arc airfoil has the maximum thickness of 4.3%,a circular average thickness of 2.5%,the maximum camber of 5.5% and an average camber of 4.5%.Numerical simulations for computing the aerodynamic performance of the airfoil were performed at the representative Reynoldsnumber between 40,000 and 100,000,and the angle of attack(AOA)from-4°to 12°by using the two dimensional steady and incompressible Navier-Stoke equations.The pressure coefficient distribution line of airfoil surfaces and the velocity vector were acquired.A carbon fiber rotor with a diameter of 40 cm,a mass of 15 g and rotor pitch of 15.7cm was manufactured with the present airfoil,and the experiments on the lift force and the structural strength in a hover state were performed and the results show the performance of the airfoil meets the use requirments.At present,the developed rotor airfoil has been successfully used in a rotor aircraft.

Aerodynamic forces on generic buildings subject to transient, downburst-type winds

Having been identified as the cause of design load winds in many parts of the world, transient winds such as gust fronts and thunderstorm downbursts have been increasingly researched over recent years. The difficulties in simulating the flow structure of downbursts in the laboratory, particularly their rapid radial acceleration and associated ring vortices, have complicated measuring wind loads on structures subject to these conditions. The University of Birmingham Transient Wind Simulator (UoB-TWS, a 1m diameter impinging jet with aperture control) has been used to simulate the transient aspects of downburst-like flow, allowing the pressure distributions they create over cube and portal framed structures to be measured for the first time, at model-scale (1:1600). Analysis of the velocity and pressure fields show that the simulator is capable of creating velocity fields which are similar to those observed in nature. Development of the ring vortex is demonstrated through phase-plot analysis. Two methods of calculating the turbulence intensity of the unsteady flow field have been used, giving mean values of between 3% and 10% depending on the method. Force coefficient time series have been estimated with the buildings angled at 0°, 45° and 90° to the radial wind direction. These are presented along with the instantaneous pressure coefficient distribution at the time of maximum roof suction. This novel research also highlights the difficulties of undertaking transient flow at model scale and drawing conclusions which are applicable to full-scale, i.e., where no two events are the same.

FSI Numerical Simulation for Wind-Induced Dynamic Response of Tension Membrane Structures

In this paper, systemic FSI numerical simulations for wind-induced dynamic response of saddle-shaped tension membrane structures are performed using ADINA software, and the effects of some parameters such as inlet velocity, wind direction, span, rise-span ratio, tension stiffness and pre-stress of tension membrane are considered. The flow is modeled with the incompressible viscous Navier-Stokes equations and the standard κ-ω turbulence model, and the membrane is assumed as isotropic-linear-elastic material with large displacement and small strain. The rules of FSI numerical simulation for wind-induced dynamic responses of saddle-shaped tension membrane structures are summarized and the wind-induced dynamic coefficients and the wind pressure distribution coefficients are obtained which can be used in the design of membrane structures.

Unsteady boundary layer measurement on an oscillating (pitching) supercritical airfoil in compressible flow using multiple hot-film sensors

Aerodynamic characteristics of an airfoil in a dynamic motion are highly affected by the behavior of the viscous boundary layer. Boundary layer condition, and the related phenomena, depends on such different parameters as the oscillation type, flight Mach number, angle of attack, etc. In compressible and transonic regimes, the shock-boundary layer interaction causes complex phenomena such as λ-shock and separation bubble on the airfoil. When oscillation and unsteady effects are also added, a complicated flow is resulted which could not be easily investigated using theoretical and computational methods. Therefore, in this paper a series of static and dynamic (pitch) tests at pre stall angle of attacks were performed in a high speed wind tunnel to study the steady and unsteady behavior of the compressible boundary layer. Measurements involved pressure coefficient distribution and qualitative analysis of output voltage of hot films. Some sinusoidal pitching motions were performed on the SC (2)-0410 airfoil at Mach numbers of 0.4, 0.5, and 0.6 with the maximum Reynolds number of 8.8 million per meter. The effects of compressibility, reduced frequency, mean angle and oscillation amplitude and free stream Mach number were investigated. Results show asymmetry between transition and relaminarization during pitching motion, and delay in transition with respect to steady conditions. Also, the location of the shock, shock induced separation bubble, and their motions due to the change of instantaneous angle of attack were detected and interpreted using the outputs of hot-film sensors.

Investigation of the internal flow field for a subsonic stage of high pressure compressor through model testing

A large-scale four-stage low-speed research compressor was designed to represent the flow field structure in a subsonic stage, which is the seventh stage of a high pressure compressor. The aim of the design work was to undertake a careful high-to-low speed transformation. The low-speed compressor was designed as an “approximate repeating-stage” configuration, in which the third stage is the model stage and is slightly different from the other stages, while the first two stages and the last stage set up inlet and outlet conditions for it, respectively. The tip–hub ratio, solidity, and aspect ratio of the low-speed research compressor were restricted to be as close as possible to those of high pressure compressor. The blade sections along the span were designed to match the distributions of high-speed surface static pressure coefficient; furthermore, stacking rules of high pressure compressor apply to low-speed blades. The low-speed blades have been tested at the 4-Stage Research Compressor Facility of Nanjing University of Aeronautics and Astronautics. Detailed traverse flow measurements were conducted in the stator blade passages of the third stage and between blade rows, along with static pressure measurements on stator blade surface and particle image velocimetry measurements in the rotor passage. Experiment results show that due to the boundary layer separated below the mid-span of the rotor blade, the total pressure loss increases at the hub region of both the rotor and stator. In addition, the experimental results in the stator blade passage reflect the characteristics of the flow field in the bowed stator passage. The results indicated that the seventh stage of high pressure compressor needs to be optimized by inhibiting the boundary layer separation mainly on the rotor blade surface below the mid-span, which is our further work.

Comparison of microburst-wind loads on low-rise structures of various geometric shapes

Microburst can produce downdraft and strong divergent outflow wind, whose characteristics are distinct from the atmospheric boundary layer (ABL) wind. In the present study, microburst-wind loading effects on low-rise structures − a cubic-shaped building, a grain bin and two gable-roofed buildings – are evaluated and compared by performing laboratory tests on scaled models using a microburst simulator at Iowa State University. Velocity and turbulence intensity profiles at selected locations were studied. The distribution of mean and root-mean-square pressure coefficients for the models are shown for selected cases and compared with those obtained in the ABL wind. Results suggest that wind loads change significantly as the radial location, orientation and geometric shape of the structures vary. It was observed that at or near the center of the microburst, high external pressures occur for all structures, resulting in a large downward force on the roof. In the outburst region, the distribution of pressure coefficients on the structure envelope was found to be similar to those in the ABL wind, although actual wind load magnitudes may be much larger in the microburst wind. Different roof slopes and its cross-section resulted in different pressure distribution and overall wind loads on the models located in the outburst region. The low-angle gable roof and conical-shaped roof experience lower drag but larger uplift in the outburst region compared to buildings with flat roof and high-angle gable roof. The geometric parameters of the roof did not influence the wind loads at or near the center of the microburst where high positive static pressures govern the wind loads. Finally, it is found that the effect of geometric scale of a model on the mean wind loads in the outburst region is minor when it is within a blockage ratio of 3% as tested in the present study.

Numerical analysis of oscillating flow around a cylinder

This paper presents the numerical model of the measuring stand - the wind tunnel, in which there is fixed a cylinder with a turbulent inlet oscillating stream. The aim of this work is the juxtaposition and comparison of characteristic values concerning the oscillating turbulent flow around the cylinder, obtained from the experiment conducted in the wind tunnel at the Institute of Thermal Machinery with the data obtained as a result of numerical modelling of unsteady phenomena. The model discussed in this paper was created using a commercial program ANSYS FLUENT that is used for mathematical modelling of flow and heat transfer processes. The expected outcome of this study is possibility of the numerical modelling of the stand concerning the analogous unsteady flows without significant invest- ment of time. Comparison of longitudinal and transverse velocity profiles in aerodynamic wake and the pressure coefficient distributions on the cylinder surface show similarities between experimental and numerical studies.

Large eddy simulation of unsteady transitional flow on the low-pressure turbine blade

The aim of this paper is to predict the phenomenon of laminar separation, transition and reattachment in a low-pressure turbine (LPT). Self-developed large eddy simulation program of compressible N-S equations was used to describe the flow structures of T106A LPT blade profile at Reynolds number of 1.1×105 based on the exit isentropic velocity and chord length. The computational results show the distributions of time-averaged wall-static pressure coefficient and mean skin-friction coefficient on the blade surface. The locations of laminar separation and reattachment points occur around 87% and 98% axial chord, which agree well with experiment data. The two-dimensional shear layer is gradually unstable along the downstream half of the suction side as a result of the spanwise fluctuation and the roll up of shear layer via Kelvin-Helmholtz (KH) instability. Three-dimensional motions appear near 84% axial chord which later triggers spanwise vortexes and streamwise vortexes, leading to transition to turbulence in the separation bubble. Through introducing the concept of dissipation function, the high loss mainly comes from the places where strong shear layer and intense fluctuation exist. Furthermore, the separation region is only an accumulation center of the low-energy fluid rather than an area of loss source.

An experimental study on the unsteady vortices and turbulent flow structures around twin-box-girder bridge deck models with different gap ratios

In the present study, an experimental investigation was conducted to characterize the unsteady vortices and turbulent flow structures around twin-box-girder (TBG) bridge deck models with and without cross beams. While the oncoming wind speed was fixed at U∞=8.0m/s (i.e., the corresponding Reynolds number, Re=1.01×104, based on the height of the box girder) during the experiments, the gap width between the TBG was varied to have four different gap ratios (i.e., the ratio of the gap width between the TBG to the deck height). The corresponding test cases were classified into two categories: the cases with relatively small gap ratios (i.e., gap ratio=0.85 and 1.70) and the cases with relatively large gap ratios (i.e., gap ratio=2.55 and 3.40). In addition to measuring the surface pressure distributions around the TBG bridge deck models using an array of digital pressure transducers, a high-resolution particle image velocimetry (PIV) system was utilized to perform detailed flow field measurements to quantify the evolution of the unsteady vortex structures around the TBG bridge deck models. The measurements reveal that as the gap ratio increases, the vortex shedding moves from the trailing edge of the leeward box to the rear edge of the windward box, which simultaneously increases the turbulent kinetic energy in the gap region and the fluctuating pressures on the leeward box. The vortex dimensions and the core-to-core distances between two neighboring vortices are also affected by the gap ratio. Combining with the estimation of the pressure field and the measured fluctuating pressure coefficient distributions on the TBG models, it is found that the TBG bridges with larger gap ratios will dramatically strengthen the fluctuating pressure coefficients on the leeward box which are greatly higher than those for the small gap ratio cases. Moreover, for large gap ratio cases, the test model with cross beams also has higher fluctuating pressure coefficients on the leeward box which just decrease a little comparing with the test model without cross beams.

Comparative Study between Field Measurement and Wind Tunnel Test of Wind Pressure on Wuhan International Stock Building

Field measurement and wind tunnel test of wind pressure on the surfaces of Wuhan International Stock Building were carried out in this paper, and the mean wind pressure coefficients, RMS wind pressure coefficients, wind pressure spectra as well as coherence functions were discussed. Meanwhile wind pressure distributions were analyzed. The results demonstrated that the distribution of the surface mean wind pressure coefficients obtained by wind tunnel test approximately agreed with that by field measurement, especially the mean wind pressure coefficients on the windward obtained by the wind tunnel test fitted those obtained by the field measurement well, while the RMS wind pressure coefficients of the wind tunnel results are smaller than those of field measurement, and the RMS wind pressure coefficients of some measure points of field measurement fluctuated greatly.

정상류 수몰 사각실린더에 작용하는 항력 특성에 관한 수치모의 연구

본 연구에서는 수치모의를 통해서 월류 흐름이 존재하는 수몰 사각 실린더의 항력 특성에 대하여 분석하였다. 모의의 신뢰성을 검토하기 위하여 실험자료와 비교하였으며 실험에서 측정하기 어려운 실린더 접촉면의 압력에 대한 분석을 통해서 상대 수심에 따른 항력의 특성을 분석하였다. 3차원 동수역학 모형을 이용한 수몰 사각 실린더의 항력 계산 결과는 실험자료의 상대 수심의 변화에 따른 항력계수의 변화를 유사하게 모의하고 있음을 확인할 수 있었다. 수치모의 결과 분석에 의하면 수몰 사각 실린더에 작용하는 항력은 대부분 압력이며 상대 수심이 증가함에 따라 전단력의 비중은 감소하였다. 실린더 접촉면의 압력계수 분석 결과에 의하면 상대 수심이 낮은 경우에는 전면부에 높은 압력계수가 형성되고 후면부에 낮은 압력계수가 형성되어 결과적으로 높은 항력계수가 나타남을 확인하였다. 상대수심이 증가하면 전면부의 압력계수는 감소하고 후면부의 압력계수는 증가하여 2차원 흐름 내의 사각 실린더와 유사한 양상을 나타낸다. 정수압 영향 분석에 의하면 전면부와 후면부의 수위 차에 의한 정수압은 항력에 미치는 영향이 제한적이며 사각 실린더에 의해 형성되는 국부적인 수위와 함께 3차원적인 흐름에 의해 형성되는 동수압의 영향이 크다는 것을 확인하였다. In this study, the drag forces on a submerged square cylinder were analyzed using a three dimensional hydrodynamic model. The numerical results were compared with the experimental results to check the reliability of the numerical simulations, and the characteristics of the drag forces with the relative depths were analyzed by analyzing the pressure acting on the cylinder surface, which are normally difficult to measure experimentally. The numerical results showed that the drag forces acting on a submerged square cylinder originate mainly from the pressure forces, and component of the shear forces decreased with increasing relative depth. The pressure coefficient distributions showed that in the case of a low relative depth, a relatively high pressure was formed in the front of a cylinder, and a relatively low pressure was formed in the rear, which gives a high drag coefficient. In a high relative depth, the pressure in the front decreased and pressure in the rear increased, which is a similar phenomenon to that normally observed in two dimensional square cylinder flow. The effect of the static pressure was analyzed and the surface elevation difference between the front and rear zone of a cylinder has a limited effect on the drag forces. Finally, the numerical results showed that the drag forces acting on a submerged square are dominated by the dynamic pressure formed by three dimensional flow and the distribution of local surface elevation.

Parametric Study of Drag Force on a Formula Student Electric Race Car Using CFD

Aerodynamic drag plays an important role in fuel economy of the vehicle especially for electric cars directly affecting the range. The objective of Aerodynamics subsystem of IIT Bombay racing team is to predict and minimize drag force on the Formula student electric race car thereby improving the performance. A standard generic car body known as Ahmed body is taken to set up simulation parameters in FLUENT by validating a test case against the experimental data available in literature. Variation and dependence of drag force on parameters such as frontal area, distribution of pressure coefficient and pressure loss in wake region is studied numerically. Comparison is made between Formula Student 2013 car Evo2 and newly designed car Evo3 for coming season of Formula Student 2014. A substantial reduction in drag force of 18.8% is achieved which can be attributed to lower frontal area and streamlined bodyworks design. Energy consumption of the vehicle for endurance race is reduced by 11.5 % improving the fuel economy.

Numerical Analysis of Separation Control Over an Airfoil Section

Flow past a rotating cylinder mounted at the leading edge of a NACA 0024 airfoil is analyzed numerically using ANSYS CFX software. The numerical simulation is done under transient conditions with shear stress transport SST turbulence model. The lift and drag coefficients are obtained and compared to the airfoil results with no rotation. The pressure coefficient distribution along the airfoil surface is also plotted along with the pressure contours. Grid independence tests are also performed to validate the results. The results indicate a delayed stall angle due to cylinder rotation in addition to an increase in the lift coefficient associated with an increase in lift to drag ratio values. The lift coefficient increases by 36 % and the stall angle is increased by 122 % for the airfoil with the rotating cylinder. All these characteristics lead to an improvement of the airfoil performance.

A study on high subsonic airfoil flows in relatively high Reynolds number by using OpenFOAM

In the present study, numerical calculations of the flow-field around the airfoil model are performed by using the OpenFOAM in high subsonic flows. The airfoil model is NACA 64A010. The maximum thickness is 10 % of the chord length. The SonicFOAM and the RhoCentralFOAM are selected as the solver in high subsonic flows. The grid point is 158,000 and the Mach numbers are 0.277 and 0.569 respectively. The CFD data are compared with the experimental data performed by the cryogenic wind tunnel in the past. The results are as follows. The numerical results of the pressure coefficient distribution on the model surface calculated by the SonicFOAM solver showed good agreement with the experimental data measured by the cryogenic wind tunnel. And the data calculated by the SonicFOAM have the capability for the quantitative comparison of the experimental data at low angle of attack.

Numerical study of turbulent wall jet over multiple-inclined flat surface

In the present work flow field of a turbulent oblique wall jet with multiple flat inclined wall has been numerically investigated. The Reynolds averaged Navier–Stokes (RANS) equations for incompressible flow are solved in transformed geometry using boundary-fitted coordinate system. The aim of the work is to investigate the influence of three inlet conditions: uniform, parabolic and trapezoidal inlet velocity profiles, on the mean flow and turbulent characteristics of the jet flowing tangentially along a multiple inclined solid wall. The results show that the inlet conditions affect the flow in the vicinity of the nozzle exit area and also the far field region. Mean flow has been presented in terms of evolution of mean velocity profiles, jet spread, decay of streamwise maximum velocity, distribution of wall pressure and skin friction coefficients, and streamwise variations of integrated momentum and energy fluxes. Discontinuities in the boundary slope of the segmented wall produces oscillations in distribution of skin friction and pressure coefficients. Reduced pressure in the region of wall discontinuity appeared to be causing the deflected jet to remain attached to the wall. The growth of the inner layer of the jet is relatively high with uniform inlet velocity profiles in the near-field region, but in the far-field region the inner layer grows with higher rate for the trapezoidal and the parabolic profiles. The rate of decay of maximum streamwise velocity for oblique wall jet on segmented wall is higher than that of plane wall jet. The jet with parabolic and trapezoidal inlet profiles show higher rate of decay than that with a uniform inlet profile. Similarity of streamwise velocity and velocity component parallel to the oblique wall has been observed in the developed region of flow. The volume flow rate of the jet varies linearly along the flow direction due to entrainment of the fluid from the surrounding. Sudden change in slope of the wall causes a sharp rise in entrained volume of fluid. The distribution of near wall velocity in the inner coordinates of the boundary layer is grossly altered and the profiles do not conform to the universal velocity profiles in the form of the law of the wall. Relatively higher position of defect law data signifies comparatively lower turbulence in the outer mixing layer. Cross-stream distribution of Reynolds stresses shows that the effect of inlet velocity profiles is more pronounced in the near field region of flow. Self-similarity is observed in the normalized turbulent shear stress profiles whereas considerable scatter can be noticed in the profiles of turbulent normal stresses for all the inlet velocity conditions. Two distinct peaks in the normalized turbulent kinetic energy profiles characterize the two shear layers of the jet and shifting of the outer peak toward the wall suggests a strong interaction of the outer layer with the inner layer as the jet develops.

Impact of transverse shear on vortex induced vibrations of a circular cylinder at low Reynolds numbers

This paper presents a numerical investigation on the vortex induced vibrations (VIV) of an elastically mounted circular cylinder in linear shear flows at low Reynolds numbers with an aim to shed light on a novel aspect of the VIV phenomena, i.e., the impact of transverse shear. In this regard, two-dimensional numerical computations are carried out by deploying a stabilized space–time finite-element formulation. The Reynolds number and the shear parameter are considered in the ranges 70⩽Re⩽500 (for a fixed reduced velocity of U∗=4.92) and 0%⩽β⩽40%, respectively. The cylinder of low dimensionless mass (m∗=10) is allowed to vibrate along both the transverse and in-line directions. The structural damping coefficient is kept zero to maximize the displacement response. Phenomena of hysteresis are observed around Re∼84 and 325. Modes of vortex shedding are 2S, C(2S) and S+P for various values of Re and β. However, only one hysteresis is observed for β=40% at Re∼84. It is further observed that the maximum displacement along the transverse direction does not get affected by the shear introduced at the inlet, however, the maximum in-line displacement depends on the shear parameter. The maximum displacement along the in-line direction increases as the shear parameter increases. For the first hysteresis (Re∼84), the extent of Re (for maximum in-line displacement) varies as the shear parameter is changed. The range of Re for the second hysteresis (for all response parameters) depends on the shear parameter such as for β=0–10% the range is 300–325, for β=20% and 30% it is 325–340 and 225–325, respectively. Strouhal number variation with Re is similar to that for other variables. Plots of pressure coefficient distribution for all shear parameters for instantaneous flow field indicate that the difference between the maximum and the minimum values of the pressure coefficient can vary significantly depending on the Reynolds number and the shear parameter.

Characteristics of failure surfaces induced by embankments on soft ground

This paper investigates the development of failure surfaces induced by an embankment on soft marine clay deposits and the characteristics of such surfaces through numerical simulations and its comparative study with monitoring results. It is well known that the factor of safety of embankment slopes is closely related to the vertical loading, including the height of the embankment. That is, an increase in the embankment height reduces the factor of safety. However, few studies have examined the relationship between the lateral movement of soft soil beneath the embankment and the factor of safety. In addition, no study has investigated the distribution of the pore pressure coefficient B value along the failure surface. This paper conducts a continuum analysis using finite difference methods to characterize the development of failure surfaces during embankment construction on soft marine clay deposits. The results of the continuum analysis for failure surfaces, stress, displacement, and the factor of safety can be used for the management of embankment construction. In failure mechanism, it has been validated that a large shear displacement causes change of stress and pore pressure along the failure surface. In addition, the pore pressure coefficient B value decreases along the failure surface as the embankment height increases. This means that the rate of change in stress is higher than that in pore pressure.