Today we are going to start a series of posts related to Computational Fluid Dynamics, CFD, addressing simple and introductory concepts as well as advanced topics based on the ICEMM’s experience in real engineering problems.
For those people, who have not had any previous contact with this technology, the Computational Fluid Dynamics, CFD, is the science that study the fluid flow, heat and mass transfer, chemical reactions and related phenomena.
What can we get from this type of analysis and simulations?
CFD simulations provide us detail information about the behaviour of fluid flow:
- Distribution of velocity, pressure, temperature, etc
- Forces exerted on solids (vehicles, aircrafts, buildings, …)
- Behaviour of fluid mixtures with other fluids or solids
- Species composition (reactions, combustions, …)
It should also be noted that CFD simulations are used in all phases of any engineering project that requires in-depth knowledge of the behaviour of the fluid, complementing testing and experimentation, and consequently reducing cost and development time.
And how does a CFD simulation work?
There are currently 2 main schemes, those based on the solution of the Navier-Stokes equations, NS, and those based on the lattice Boltzmann method, LBM.
The first is the traditional scheme, with a long history and validation in a large number of fluid mechanics and thermal problems. It is used by leading CFD software such as Ansys Fluent, Star CCM+, MSC Cradle or Acusolve. It should be noted that the solutions of the Navier-Stokes equations can be done by applying the Finite Volume technique (Ansys Fluent, Star CCM+ or MSC Cradle) or by Finite Element (Acusolve or Comsol). The advantages and disadvantages of both methods are well known and will be discussed ins subsequent publications.
Regarding the lattice-Boltzmann method, it is important to point out the advantages of its use in the pre-processing phase, where the generation of the volume to analysed is almost automatic regardless of the complexity of the geometry, as well as the treatment of turbulence and transient processes, especially with moving geometries. The main disadvantages of this method as present are the computational requirements, the impossibility to evaluate stationary processes efficiently and the lack of validation cases compared to the Navier-Stokes based method.
In ICEMM we have only worked with the method based on Navier-Stokes so far, so that all information hereinafter will be related to this method.
The solution of the Navier-Stokes equations
The operation of a CFD simulation based on solving the Navier-Stokes equations can be summarized in the following steps:
- Discretisation of the domain to be studied into a finite number of control volumes or cells
- Solving the conservation equations of mass, momentum, energy, species, … in the set of control volumes
- Discretisation of the equations in partial derivatives in a system of algebraic equations and their resolution
The application of the above steps in solving any CFD simulation can be summarised according to the following scheme, which applies to any of the CFD software currently in use:
The next posts will be focused on explaining how we apply the above scheme to different type of problems with Acusolve and FDS software. The posts that will be published in the blog of our website and in the download area will be the following ones:
- Introduction to CFD simulation:
- 1_1_Acusolve_Workflow – Basic Tutorial – Interface (Acusolve)
- 1_2_Vortex_Shedding – Transient Analysis (Acusolve)
- 1_3_Fire_in_Garage – Mulitphase Analysis (FDS)
- 1_4_Cooling_Device – Boundary Layer Analysis (Acusolve)
- 1_5_Backward_Step – Turbulence (Acusolve)
- 1_6_Turbulence_Grid – Turbulence (Acusolve)
- 1_7_Airfoil – Compressible Flow (Acusolve)
- 2_1_Curved_Channel – SST with Curvature Correction
- 2_2_Diffuser – Turbulence RANS Models
- 2_3_Airfoil – Laminar to Turbulence Transition
- 2_4_Stable_Pipe – LES Simulation
- 2_5_Periodic_Hills – Hybrid RANS-LES models
- 2_6_Cylinder – Hybrid RANS-LES + Transition effects
- 2_7_Bluff_Body – Scale-Adaptive Simulation (SAS)