Air Pollution within a Street Canyon — ANSYS Fluent CFD SimulationDescriptionThis project simulates pollution diffusion in a street canyon using ANSYS Fluent. When two rows of building blocks stand parallel to each other, the space between them forms what is known as a street canyon (or urban canyon) — a configuration whose geometry strongly influences how urban heat and airborne gases are distributed. This project investigates the amount and distribution of pollutants within that canyon space. Within the Urban Microclimate: All Levels CFD Training Package, this project introduces pollutant dispersion through the classic street-canyon geometry, applying the Species Transport model to the fundamental unit problem of urban air quality.MethodologyThe three-dimensional model is designed in Design Modeler, with a computational area 36 m long, 24 m wide, and 8 m high containing two rows of simple building blocks parallel to each other. To reduce computational cost, the model is limited in extent and symmetry boundary conditions are applied around the urban area. Meshing in ANSYS Meshing produces 1,938,659 elements. The Species Transport model is used to represent the two gaseous species — air and pollutant — where the pollutant has a specific heat capacity of 1100 J/kg·K and a molecular weight of 77.49064 kg/kmol, and air has 1006.43 J/kg·K and 28.966 kg/kmol. All pollutants are assumed to be generated within the canyon: two grooves in the ground act as the pollution source, with a source term of 0.011 kg/m³·s. Initially only air fills the domain, and pollutants then begin to be produced. At the inlet, pure airflow enters through a velocity-inlet condition whose magnitude varies with the inlet location — implemented as a profile in UDF format — with the air temperature set to 300 K. The RNG k-epsilon model and the energy equation are enabled to resolve the turbulent flow and the temperature field.AnalysisAt the end of the solution, three-dimensional contours of pressure gradient, velocity, temperature gradient, air mass fraction, and pollutant mass fraction are obtained, along with two-dimensional contours of velocity, air mass fraction, and pollutant mass fraction. The results show air pollution originating from the interior of the street canyon, and the two- and three-dimensional velocity vectors reveal a vortex — a rotation of the flow — forming inside the canyon, which governs how the pollutant is trapped or cleared. By the end of this project, you'll be able to set up a Species Transport simulation with a ground-level pollution source, apply a UDF velocity profile at the inlet, and interpret the concentration and velocity fields that characterize air quality in a street canyon.

Non-Newtonian Blood Pulsatile Flow in a Vein — ANSYS Fluent CFD Simulation TrainingThis project simulates non-Newtonian, pulsatile blood flow through a vein using ANSYS Fluent, with the full case analyzed through CFD post-processing.The working fluid is blood, a non-Newtonian fluid. Non-Newtonian fluids are those whose viscosity changes with shear rate, meaning they have no single fixed viscosity. In such fluids the relationship between shear stress and applied strain rate is nonlinear, so no constant viscosity coefficient applies. The simulation is run as transient over 0.5 s, and a User-Defined Function (UDF) is applied to model the pulsing of the blood flow. Because blood flow is not steady but pulsed, the velocity is prescribed as a periodic function through the UDF code.The geometry was created in Gambit. The model consists of a main cylindrical vessel and two smaller branch vessels of reduced size and diameter — one branching at a 90-degree angle and the other with a 45-degree curvature. It has one inlet section and two outlet sections.Meshing was performed in ANSYS Meshing using an unstructured grid, for a total of 397,388 cells.MethodologyThe working fluid is blood, with a density of 1050 kg/m³. Because blood is non-Newtonian, its viscosity is described using the Carreau model with appropriate parameters.Newtonian fluids maintain a constant viscosity under applied force, whereas non-Newtonian fluids exhibit variable viscosity, of which there are several types. Time-dependent non-Newtonian fluids fall into two categories: rheopectic fluids, such as printer ink and cream, whose viscosity increases over time under load, and thixotropic fluids, such as honey, whose viscosity decreases as force is applied. Time-independent non-Newtonian fluids divide into three groups: dilatants, such as starch and clay, whose viscosity depends only on the magnitude of the applied force; pseudoplastics, such as greases, paints, soaps, and ketchup, whose viscosity is inversely related to the applied force; and Bingham fluids, such as toothpaste and silica nanocomposites, which require a threshold stress before they begin to flow.In this simulation, blood is treated as a pseudoplastic non-Newtonian fluid defined by the Carreau model. This model spans a wide range of fluid behavior by fitting a curve that matches both Newtonian and shear-thinning (pseudoplastic) responses.ResultsAfter the solution is complete, contours of pressure and wall shear stress are obtained at several time instants. The results confirm that the flow inside the vessel is fully pulsatile, since the pressure varies over time. They also show that pressure and wall shear stress are correlated: as the pressure inside the vessel rises, the wall shear stress increases accordingly.

Aorta, Non-Newtonian Pulsating Blood Flow — ANSYS Fluent CFD SimulationDescriptionThis project studies non-Newtonian pulsating blood flow in the aorta using ANSYS Fluent. The aorta geometry is obtained from a real CT scan, provided as an STL file that must be repaired before meshing — a workflow representative of patient-specific biomedical CFD. Blood is a non-Newtonian fluid whose apparent viscosity changes with shear rate, and the aorta's pulsatile flow, curvature, and branching make it a rich, realistic case for studying how such a fluid behaves in a large vessel. Within the Non-Newtonian Flow: Beginner CFD Training Package, this project builds on the earlier blood-flow case by moving to a larger, geometrically complex vessel reconstructed from real medical imaging.MethodologyThe aorta geometry is obtained from a CT scan, and tools such as SpaceClaim, ICEM CFD, and Design Modeler can be used to repair it; here ICEM CFD was used to fix the geometry and generate the mesh. The mesh was first generated with the octree method using five layers of prism cells at a ratio of 1.2, then improved with the Delaunay method, giving a final count of 457,864 cells. A UDF defines the pulsatile inlet velocity. The non-Newtonian behavior of blood is captured with the Carreau model, in which viscosity depends on the shear rate, defined by the zero-shear viscosity (µ₀), the infinite-shear viscosity (µ∞), the power index (n), and the relaxation time (λ). No energy equation is included, so temperature is neglected. The solver is transient, the flow is turbulent, and the density is constant at 1060 kg/m³, with a no-slip condition on the inner surface of the vessel wall. The UDF used to define the pulsating inlet velocity is provided.AnalysisThe results illustrate the inlet velocity and pressure drop over the pulse cycle, with the maximum velocity occurring at 0.15 s. The wall shear stress (WSS) contours show the maximum values in the aorta sections of smaller diameter, while the static-pressure contours show that at the beginning of the blood pumping, the pressure is highest at the entrance of the branches. When suction occurs at 0.4 s, it has the greatest impact on the inlet section of the aorta. Animation files of pressure and shear stress are included to reveal the pulsatile behavior and give a clearer understanding of the flow. By the end of this project, you'll be able to repair a real STL geometry from medical imaging, generate a prism-layer mesh, apply the Carreau non-Newtonian model with a UDF-defined pulsatile inlet, and interpret the velocity, pressure, and wall-shear-stress fields that characterize pulsatile blood flow in the aorta.

DescriptionIn this project, we present a simulation of a Blood Vessel via ANSYS Fluent software.Since the vessel is exposed to blood flow, an interaction occurs between the blood flowing and the vessel structure. First, the blood flow exerts a force on the vessel's body by hitting it. Subsequently, displacement or deformation appears on the vessel, which can lead to the blood flow being affected. Therefore, we intend to perform a numerical simulation of the blood vessel as a Fluid-Structure Interaction (called FSI).The interaction between fluid and structure can be implemented as:One-way FSITwo-way FSIIn this project, we aim to analyze both the effect of fluid on the structure and the effect of the structure on the fluid. So, we choose Two-way FSI, which is a more accurate and realistic but more complex approach.We modeled the geometry via Spaceclaim software. The computational domain is a sample space of a vascular system with a simple construction. We considered the blood vessel as a horizontal cylinder with a solid layer surrounding the fluid region.We meshed the computational domain via ANSYS Meshing software. The mesh is of an unstructured type, and approximately 56,000 cells have been generated.MethodologyFluid-structure interaction can be performed in two general methodologies:In the ANSYS Workbench environment, using an external solver (specifically, system coupling)Only in the Fluent solver (in the form of an intrinsic FSI).In this project, we implemented a two-way FSI in the ANSYS Fluent environment. In other words, the Fluent solver performs both fluid and solid calculations simultaneously.For two-way FSI in Fluent solver, the Structure model is utilized. The structural model can be implemented in two ways:Linear elasticity: The deformation is proportional to the applied force. In this case, the deformations are usually small, and the calculation process is faster.Nonlinear elasticity: The deformation is not necessarily proportional to the applied force. In this case, the deformations are usually large, and the calculation process is more complex and time-consuming.In this project, we considered fluid-structure interaction in the form of a Linear Elasticity state.Since we were analyzing two-way FSI and considering the effect of structural displacement on the adjacent fluid, we used the Dynamic Mesh model. In other words, we establish a connection between the fluid and structural calculations with the Intrinsic FSI option. Then, we enabled the smoothing and remeshing methods to define a deformable mesh.In addition, for defining blood flow in a pulse-mode, we used a user-defined function (UDF) so that the flow has a variable velocity with respect to time.ResultsWe analyzed the results in two fluid and solid approaches:In a fluid view, we studied the behavior of blood flow. For this, we obtained the distributions of the pressure and velocity of blood. The results show that the blood flow collides with the vessel body at pulsatile speed and, as a result, exerts a hydraulic force on the vessel structure.In a solid view, we studied the behavior of the vessel body under the influence of the applied forces of the blood flow. For this, we obtained the distribution of the von Mises stress and displacements (in all directions). The results confirm that the blood flow affects the vessel structure and, as a result, it undergoes deformation relative to the initial state.In conclusion, we can claim that we carried out the simulation project of a blood vessel correctly and acceptably by using the two-way FSI method.

Lumen Blood Vessel (Non-Newtonian) — ANSYS Fluent CFD SimulationDescriptionThis project simulates a lumen blood vessel using coupled Fluid-Structure Interaction (FSI) together with a non-Newtonian blood model in ANSYS Fluent. Because blood is a shear-thinning fluid whose viscosity changes with the local strain rate, a non-Newtonian treatment is essential for capturing the flow behavior realistically inside the vessel — and because the elastic vessel wall deforms under the pulsating flow, the case couples the fluid and structural response. Within the FSI: Beginner CFD Training Package, this project builds on the pulsatile blood-vessel case by adding non-Newtonian blood behavior, giving a more physically realistic biomedical FSI problem.MethodologyThe three-dimensional geometry was created in SpaceClaim, with a computational domain 164 mm long, 262 mm high, and 5 mm wide, meshed in ANSYS Meshing to a total of 356,794 elements. Owing to the pulsatile nature of the problem, a transient solver was used. A blood vessel together with its wall is simulated in ANSYS Fluent, with the solver's intrinsic FSI module enabled so that the displacement of the vessel wall could be captured in response to the flow. The inlet boundary condition was defined as a pulsatile velocity through a UDF, while the outlet was defined as a pulsatile pressure, also supplied through a UDF. The blood itself was modeled as a non-Newtonian fluid using the Carreau model, which reproduces the shear-thinning drop in viscosity as the shear rate increases, and a laminar model was enabled to solve the fluid equations.AnalysisOn completion of the solution, three-dimensional contours of wall displacement and von Mises stress were obtained. As the results show, the blood flowing through the vessel exerts stress on the vessel walls, deforming them and demonstrating the two-way coupling between the pulsatile non-Newtonian flow and the compliant vessel structure. From these results you can evaluate how the pulsating blood loads the vessel wall, where the stress and deformation concentrate, and how the shear-thinning viscosity shapes the flow. By the end of this project, you'll be able to set up a coupled FSI simulation with a compliant vessel wall, apply the Carreau non-Newtonian model with UDF-defined pulsatile inlet and outlet conditions, and interpret the wall-displacement and von Mises stress fields that characterize biomedical fluid-structure interaction.

DescriptionThis project simulates a non-return (check) valve, a device that allows flow in one direction while blocking reverse flow, using ANSYS Fluent. Such valves are needed wherever downstream pressure can rise above inlet pressure, since without them the flow would push backward through the system and potentially damage it. The valve motion is captured through dynamic mesh with one-degree-of-freedom rotation, letting the valve flap swing open and closed in response to the flow rather than following a prescribed motion. The inlet velocity is driven by a UDF that ramps up to 1 m/s over the first 0.4 seconds, then drops to a near-zero value of 0.000001 m/s afterward, simulating an abrupt loss of driving flow. The geometry is a 26 cm × 5 cm two-dimensional domain built in SpaceClaim and meshed in ANSYS Meshing with an unstructured grid of 61,580 elements.MethodologyThe valve dynamics are handled through the Six DOF solver with only one rotational degree of freedom enabled, and a spring stiffness of 1 N·m/rad is added to help drive the valve closed once the flow subsides. Turbulence is modeled with SST k-omega, and the solution uses a transient, pressure-based solver with gravity neglected. The inlet is a velocity-inlet driven by the UDF profile, the outlet is a pressure-outlet at 0 Pa gauge, and all other walls are stationary. Pressure-velocity coupling uses SIMPLE, with second-order discretization for pressure and momentum, and first-order upwind for turbulent kinetic energy and dissipation rate; initialization is standard.AnalysisThe results show the valve opening for the first 0.4 seconds while the high-velocity, high-kinetic-energy inflow pushes it open, then beginning to close as soon as the flow velocity drops toward zero. The spring force reinforces this closing motion, accelerating valve closure and ensuring the flow cannot slip back through the inlet once the driving pressure is gone. The reported UDF velocity-versus-time profile and the force-on-valve-versus-time plot together show this open/close cycle directly, tying the valve's mechanical response to the imposed flow transient.

DescriptionThis project simulates a floating vessel's motion on water using the dynamic mesh method in ANSYS Fluent. The vessel is positioned at the center of a three-part computational domain, designed so its center of gravity sits along the vertical axis for simulation convenience. The geometry is built in 3D in Design Modeler, and meshed in ANSYS Meshing with an unstructured grid near the vessel and a structured grid elsewhere, totaling 902,808 elements.MethodologyBecause the vessel's motion requires the mesh to deform continuously around it, the Dynamic Mesh model is used, combining smoothing, which adjusts mesh boundaries without changing node count or connectivity, with remeshing, which reconstructs cells that become too distorted when boundary displacement is large relative to local cell size. The domain is divided into a small moving zone around the vessel, a surrounding deforming zone, and a larger stationary outer zone, with the vessel and its moving zone treated as a rigid body via the Six Degrees of Freedom (6-DOF) model. Since the vessel is physically constrained to only vertical translation and rotation about its central axis, a UDF restricts the 6-DOF motion down to these two degrees of freedom, with the vessel's center of gravity and rotation axis specified explicitly in the rigid body setup. The water and air phases are captured with the VOF multiphase model, air above and water below, both entering horizontally at 1.44 m/s and exiting at atmospheric pressure, with an Open Channel boundary condition at the outlet defining the water level. Given the fundamentally time-dependent nature of dynamic mesh motion, the simulation runs transient, covering 7 seconds at a 0.01 second time step.AnalysisThe results include 2D pressure contours on the vessel and 2D velocity and volume fraction contours in the surrounding air-water region, taken at the final second of the simulation, along with time-history plots of the vessel's vertical displacement and rotation angle over the full 7 seconds. These plots show the oscillation amplitude in both displacement and rotation decreasing over time, with the vessel's motion becoming effectively damped by the seventh second. At that point, the vessel settles near a vertical position of z = 0.021 and a rotation angle of Y_theta = -1.338, indicating it reaches a stable floating equilibrium consistent with the physical damping expected in this kind of fluid-structure interaction.

DescriptionIn this project, we present a simulation of an Airfoil exposed to the airflow via ANSYS software.Since the airfoil is exposed to airflow, an interaction occurs between the wind blowing and the airfoil structure. First, the airflow exerts a volume force on the airfoil's body by hitting it. Subsequently, displacement or deformation appears on the airfoil, which can lead to the airflow being affected. Therefore, we intend to perform a numerical simulation of the airfoil as a Fluid-Structure Interaction (called FSI).The interaction between fluid and structure can be implemented as:One-way FSITwo-way FSIIn this project, we aim to analyze both the effect of fluid on the structure and the effect of the structure on the fluid. So, we choose Two-way FSI, which is a more accurate and realistic but more complex approach.We modeled the geometry via Design Modeler software. The computational domain is a sample space of the surrounding air that includes both fluid and solid domains. There is a solid airfoil structure within the fluid environment, which is considered fixed from the center.We meshed the computational domain via ANSYS Meshing software. The mesh is of an unstructured type, and approximately 56,000 cells have been generated.MethodologyFluid-structure interaction can be performed in two general methodologies:In the ANSYS Workbench environment, using an external solver (specifically, system coupling)Only in the Fluent solver (in the form of an intrinsic FSI).In this project, we implemented a two-way FSI in the ANSYS workbench environment.For two-way FSI with an external solver, three main steps are required:Simulation of the fluid domain from the model using the Fluent solverSimulation of the solid domain from the model using the Transient Structural solverDefinition of the Data Transfer between the fluid and structural solvers using the System Coupling toolFor utilizing the system coupling, we define two data transfers:In the form of Forces to the interface wall (from the fluid solver to the structural solver)In the form of Displacements of the interface wall (from the structural solver to the fluid solver)Since we were analyzing two-way FSI and considering the effect of the structure's displacement on the adjacent fluid, we used the Dynamic Mesh model. In other words, we establish a connection between the fluid and structure calculations with the System Coupling option. Then, for defining a deforming mesh, we enabled the smoothing and remeshing methods.In addition, because of the aerodynamic nature of the airfoil and the very high airflow velocity, we considered a density-based solver.ResultsWe analyzed the results in two fluid and solid approaches:In Fluent, we studied the behavior of airflow. For this, we obtained the distributions of the pressure and velocity of air. The results show that the airflow collides with the airfoil body at high speed and, as a result, exerts a hydraulic force on the airfoil structure.In Structural Transient, we studied the behavior of the airfoil body under the influence of the applied forces of the airflow. For this, we obtained the distribution of the deformation, von Mises stress, and elastic strain. The results confirm that the airflow affects the airfoil structure and, as a result, it undergoes displacements relative to the fixed center.In conclusion, we can claim that we carried out the simulation project of an airfoil correctly and acceptably by using the two-way FSI method.

Spiral Magnetic Separator CFD Simulation Using ANSYS FluentIntroductionThis study investigates the performance of a spiral magnetic separator using computational fluid dynamics to understand the complex interactions between fluid flow, magnetic particles, and an applied magnetic field within the separator. Water enters the domain from the upper boundary carrying both magnetic particles and SiO2 particles, while an applied magnetic field, represented through user-defined functions for the Bx, By, and Bz components, influences the trajectory of the magnetic particles and enables their separation from the non-magnetic SiO2 particles. By combining turbulence modeling, discrete phase modeling, and magnetohydrodynamics, this research provides valuable insight into the separation efficiency and flow behavior characteristic of magnetic separation systems.Geometry and MeshThe geometry consists of a spiral-shaped separator with multiple turns, designed in ANSYS SpaceClaim and meshed in ANSYS Meshing to promote effective particle separation along the spiral flow path. The simulation was conducted using a steady-state, pressure-based solver in ANSYS Fluent to capture the coupled flow and particle behavior throughout the domain.MethodologyTurbulent flow within the separator was resolved using the Realizable k-epsilon model with standard wall functions. A two-way coupled Discrete Phase Model was implemented to simulate the behavior of both magnetic and SiO2 particles, capturing the interaction between the particles and the continuous water phase. Group injection was defined for both particle types, with diameter distributions specified using the Rosin-Rammler model. The Magnetic Induction MHD method was enabled with a DC field type to simulate the effects of the applied magnetic field on both the flow and particle trajectories, with several user-defined functions implemented to define the source terms for the Bx, By, and Bz magnetic field components.Results and ConclusionThe magnetic field components exhibit alternating positive and negative regions along the spiral path, with By ranging from -1.5347×10⁻¹⁵ to 1.7298×10⁻¹⁵ T, Bz ranging from -1.0879×10⁻¹⁴ to 8.7105×10⁻¹⁶ T, and Bx displaying a more complex distribution between -3.79×10⁻¹⁵ and 4.40×10⁻¹⁵ T. Static pressure within the separator ranges from -0.84606 to 4.6414 Pa, with higher pressures concentrated near the outer walls of the spiral, while velocity magnitude varies from 0 to 0.13735 m/s, with higher velocities observed near the inner walls. Particle tracks reveal a polydisperse mixture with diameters ranging from 1.00×10⁻⁴ to 2.96×10⁻⁴ m, experiencing static pressures between -8.94380 and 9.69623 Pa as they travel through the domain. Pathlines colored by Bx and velocity magnitude illustrate the complex spiral flow pattern, with velocities along the pathlines reaching up to 0.181 m/s in the upper turns of the spiral. The particle tracks further indicate a gradual separation of particles based on their magnetic properties and size, with larger and more strongly magnetic particles tending to concentrate toward the outer walls of the spiral. These results confirm the effectiveness of the spiral design in creating an extended separation path, where the combined variation in magnetic field strength and flow velocity along the spiral drives progressive particle segregation based on each particle's position within the separator.

DescriptionThis project uses ANSYS Fluent to simulate the flow of a nanofluid through a solid aluminum channel under an applied magnetic field. The flow is steady and modeled as a single-phase flow, with the thermophysical properties of the nanofluid — density, viscosity, specific heat, and thermal conductivity — calculated as functions of the nanoparticle volume fraction. The core of the study is the magnetohydrodynamic (MHD) interaction: the applied magnetic field acts on the electrically conducting nanofluid, altering its flow and heat-transfer behavior through the Lorentz force and Joule heating. The surface-averaged temperature of the nanofluid rises from 293.2 K at the inlet to 304.175 K at the outlet.Geometry & MeshThe fluid domain was created in SpaceClaim, and the computational grid was generated in ANSYS Meshing. The mesh is unstructured, with 26,000 elements.MethodologySeveral assumptions underpin the simulation: a pressure-based solver is used, the formulation is steady, and gravitational effects are neglected.Models — the energy equation is enabled; turbulence uses the standard k-epsilon model with standard wall functions; and the MHD model is applied using the magnetic-induction method, solving the MHD equations with the Lorentz force and Joule heating both included.Magnetic field — an external field B₀ is imposed by patch, with a 1 T component applied in the relevant directions.Materials — the working fluid is a water-based nanofluid (density 1312 kg/m³, specific heat 3248 J/kg·K, thermal conductivity 1.09387 W/m·K, viscosity 0.0011 kg/m·s, electrical conductivity 1,000,000 S/m, magnetic permeability 1.257 × 10⁻⁶); the solid channel is modified aluminum (density 2719 kg/m³, specific heat 871 J/kg·K, thermal conductivity 202.4 W/m·K, electrical conductivity 3.541 × 10⁷ S/m). The solid also carries an energy source representing Joule/MHD heating of 1,000,000 W/m³ applied through a UDF.Boundary conditions — Inlet: velocity inlet at 1 m/s, 5% turbulence intensity, turbulent viscosity ratio 10, and 293.2 K; the outer solid wall is held at 320 K (insulating for the magnetic field), and the fluid-solid interface is a coupled wall.Methods — SIMPLE pressure-velocity coupling; least-squares cell-based gradients; second-order for pressure, momentum, and energy; and first-order upwind for the turbulence quantities and the magnetic field components. The solution is initialized with a 1 m/s velocity and a temperature of 293.2 K.ConclusionWithout a magnetic field, the nanofluid's average temperature rises from 293.2 K at the inlet to 304.175 K at the outlet. When the magnetic field is applied, the outlet temperature increases further to 305.14 K. Plots of temperature and velocity along the centerline of the domain are presented for both cases (with and without MHD).Comparing the outlet temperatures with and without the magnetic field reveals the effectiveness of the MHD effect in this problem: applying the field raises the outlet temperature by about 1 K. This demonstrates how a magnetic field, through the Lorentz force and Joule heating acting on a conductive nanofluid, can be used to enhance heat transfer — the central principle behind MHD-based thermal management.