Computational Fluid Dynamics

Computational fluid dynamics (CFD) is the analysis of systems that involve fluid flow, heat transfer and associated phenomena by a set of numerical methods simulating the equations that govern fluid motion. In effect, CFD combines engineering fluid dynamics, heat transfer, and mathematics with computer science. CFD analysis plays a very important role in systems from an engineering perspective. For example, let us assume that engineers need to build a bridge across a river. The river is located in India, and receives heavy rains in the monsoon season. Secondly, heavy military vehicles regularly ply the bridge as it is the only bridge that connects two military posts across the river and is very vital for the defence of India. What should be the shape and strength of pillars so that the bridge sustains not only eddies that arise in flood like situations of the monsoon, but also the weight of the heavy military vehicles? Let’s take another example. Let’s say an aerospace company wants to build a new aeroplane. What should be the lift and drag of the aeroplane so that it flies smoothly under all conditions? It is important to remember that the aerospace companies also need to be aware of prevailing regulations, financial implications and environmental impact while designing for safety. Since aeroplanes also face air turbulence (remember that air is a fluid) while flying through rough weather, that is another consideration. CFD analysis is one of the prime considerations while designing aircrafts.

These are only two examples that underline the importance of computational fluid mechanics. Other examples where CFD analysis is critical include:

• Hydrodynamics of ships and marine engineering
• Internal combustion engines and gas turbines in a power plant
• Chemical process industry like agitators and mixers
• Distribution of pollutants and effluents in chemical engineering
  

Basics of Computational Fluid Dynamics
The physical aspects of any fluid flow are governed by the three laws of Newton. In essence, these three laws talk about the conservation of mass, momentum and energy. These fundamental principles can be expressed in terms of mathematical equations, which in their most general form can be represented as partial differential equations. Computational fluid dynamics is in part the art of replacing the governing partial differential equations of fluid flow with numbers, and advancing these numbers in space and / or time to obtain a final numerical description of the complete flow field of interest.  CFD assumes that the density of the fluid under consideration is high enough so that it is always under continuum. This implies that even an infinitesimally small element of the fluid can be represented using differential calculus. Using this assumption, we are able to define velocity, pressure, temperature, density and other important quantities at point of the fluid.

Let us briefly delve into each discipline that defines computation fluid dynamics. Fluid dynamics deals with fluids. As far CFD is concerned, we are mostly concerned with fluid that is in motion, and how it behaves when flowing.  The physical characteristics of this fluid flow can be expressed by complex mathematical equations, and they are converted into numeric simulation by computer scientists using high-level computer programming languages into computer software packages. One of the best known, industry benchmark software is Altair AcuSolve™.

All CFD codes contain three main elements:

1. A pre-processor, which is used to input the problem geometry, generate the grid, and define the flow parameter and the boundary conditions to the code.
2. A flow solver, which is used to solve the governing equations of the flow subject to the conditions provided. There are four different methods used as a flow solver:finite difference method
   
   • finite difference method
   • finite element method
   • finite volume method
     
3. A post-processor, which is used to massage the data and show the results in graphical and easy to read format.

As you can now understand, CFD is a complex topic requiring huge amount of computation. In the past, when computes were not so powerful or prevalent, it was the privy of high technology engineering areas of aeronautics and astronautics. It is now routinely used in solving complex problems in modern engineering practice, thanks to the emergence of high speed computers. CFD solutions generally require repetitive manipulation of millions of numbers, something that cannot be accomplished without high computer power. The aerospace industry has integrated CFD analysis in their R&D, as has the automobile industry and the civil industry. CFD finds uses in all walks of life, and thanks to refined software like Altair AcuSolve^®, it is no longer the privy of the elite few.

Computational Fluid Dynamics Services
As CFD software grows more and more advanced (read: capable of solving complex differential equations), tremendous computing power is needed to really leverage their capabilities. CFD simulations rely on high performance computing (HPC) to solve more complex, more real, more detailed, and bigger problems. Having good software is part of the solution, but interpreting the results is a different ballgame altogether.

A CFD simulation can be divided into four main phases: (i) mesh generation, (ii) setup, (iii) solution and (iv) analysis and visualisation. Ideally, these processes should be carried out successively. In reality, they are interleaved as there are many variables involved in CFD simulation and their interdependency is quite complicated. In addition, CFD does not provide black and white answers; it takes experience to interpret the results correctly. Turbulence and combustion modelling, heat transfer, efficient solution techniques for viscous flows, etc. all require intelligent interpretation. Mesh adaptation is one of the key technologies needed in CFD that is tricky to handle.  Many manufacturers do not have either HPCs or the skill-set to interpret the CFD results accurately. This is where CFD service providers (like DesignTech Systems) step in. Since they specialize in providing many engineering services - one of which is CFD – they have enough expertise to carry out CFD simulation properly. More importantly, for small manufacturers who do not want to invest heavily in CFD software, these services are a perfect solution.

CFD analysis is critical, and indeed may be the only choice in certain instances. Suppose structural engineers need to erect a dam across a river. The catchment area is very large; the dam will provide not only water to nearby villages but also generate and supply hydroelectricity to the grid. Engineers cannot just build a trial dam across the river, as that will entail a huge cost. In addition, it will be time consuming as well. Clearly, a controlled experiment is not viable in this answer, and engineers need to rely on precise CFD analysis of the water flow to devise the proper solution.

Summary
Computational fluid dynamics is a very crucial aspect of engineering simulation and analysis. It is complicated in nature, and few manufacturers have both the resources and the expertise to carry it properly. Outsourcing CFD analysis to trusted engineering service providers is the right answer for most manufacturers.