SIMULATION SCIENCE and COMPUTATIONAL FLUID DYNAMICS
Posted October 5, 2015 at 06:13 by adarsh tiwari
The wheel invented centuries ago has now taken speed. World today is fast moving with new product extensively coming in to replace the existing product. Experimentation is always necessary to introduce a product. Only experiments can validate the theory and purpose of a product. However, experimentation is not very easy and cheaper. The increasing cuttingthroat competition in the world is driving the business leaders to find a new and cheaper way to assist, if not replace, the need of experimentation.
We humans have invented mathematics as a solution to the problems of the physical world. And we are struggling to solve various equations to get the proper solutions to the real life problems. This same is applied everywhere in engineering product design and development. As I mentioned earlier mathematics will closely approximate the reallife problem and hence we are solving the corresponding equations. But these equations are complicated and we do further approximations to simplify our efforts. We all know life is not simple, so is the equations. Hence, it is least possible for any human to solve the equations for the complicated situations like a fluid flow say blood in our bodies, water in a steampower plant pipelines, equations for conversion of water to steam and what not.
We are surrounded with fluids everywhere, from the air we breathe to the blood flow in our body, from water in domestic pipelines to the steam in the thermal power station. The estimate of fluid flow and flow parameters are very important to develop an upgraded product. For example just take the first aircraft of modern days, made by Wright brother and compare it with the latest available aircrafts. Present day aircrafts are much faster, efficient and way better than the first built.
But was the first one unimportant?
Absolutely not. Actually, it was our first step to the better and efficient aircrafts. Our aerofoil structures are designed mush better these days. But have you ever imagined the cost of experimentation for these newly designed aerofoil?
Depending upon the experiments the cost may range from few lacs to several crores. What if these experiments fail? Absolutely the first thing is direct loss to that amount. Secondly, there is an advantage to the competitor because he had already saved that money.
So the question is, Is there something which we can do to save this money?
The answer is yes and this lies in the description above. As mentioned above, we are closely approximating the physical phenomenon using mathematical equations but we are facing difficulties in solving them. So, it would be better if we generate the equations and let some machine compute it. This is the mother thought of invention and development of simulation science. This science is still under development phase but the progress is really astonishing. Now we can simulate from the stresses to the complicated flow using computers.
One of the simulation science deals extensively with the fluid flow and is named as ‘Computational Fluid Dynamics’ globally adopted as abbreviation of CFD. By seeing the beautiful pictures obtained in the results, some people also call it as Colourful Fluid Dynamics .
For me, it is a combination of both art and science. I can surely justify it for you. Most of the people reading this blog are comfortable with MS Word, but how many of us are a good writer?
There is a mechanic in a garage who knows how to use a drill and an engineer is also there. Both know how to drill but does the mechanic knows where to drill? Similarly, one may say a mathematician and a computer graduate can together solve an equation using computer. The mathematician will write the algorithm to solve while the computer science fellow will write the program for the same. But can we say confidently that they understood the physics of the problem? The results they obtained using CFD tools are correct? Do the results make sense?
In fact, if there is improper boundary conditions, or improper grid or even improper method to solve the equations, the results obtained will be erroneous. I will be similar to using a drill for tightening a screw.
There are a lot of solvers commercially available like Ansys Fluent, CFX, OpenFOAM StarCCM etc. But we need a proper understanding of the subject before proceeding toward the subject. Here in Bangalore I have seen a lot of training institutes claiming to train in CFD. Personally I have tested few of them but unfortunately there is only one training institute which gives proper knowledge and guidance in CFD. It is not like I am doing advertising for them but seriously the way MyCADCFD () handles the concepts and apply it the software tools is really astonishing and interesting. They said they are having two branches in Pune and Bangalore. They also do online training.
In further blogs, we will talk about some interesting cases and results of CFD.[/LEFT][/RIGHT][/RIGHT]
We humans have invented mathematics as a solution to the problems of the physical world. And we are struggling to solve various equations to get the proper solutions to the real life problems. This same is applied everywhere in engineering product design and development. As I mentioned earlier mathematics will closely approximate the reallife problem and hence we are solving the corresponding equations. But these equations are complicated and we do further approximations to simplify our efforts. We all know life is not simple, so is the equations. Hence, it is least possible for any human to solve the equations for the complicated situations like a fluid flow say blood in our bodies, water in a steampower plant pipelines, equations for conversion of water to steam and what not.
We are surrounded with fluids everywhere, from the air we breathe to the blood flow in our body, from water in domestic pipelines to the steam in the thermal power station. The estimate of fluid flow and flow parameters are very important to develop an upgraded product. For example just take the first aircraft of modern days, made by Wright brother and compare it with the latest available aircrafts. Present day aircrafts are much faster, efficient and way better than the first built.
But was the first one unimportant?
Absolutely not. Actually, it was our first step to the better and efficient aircrafts. Our aerofoil structures are designed mush better these days. But have you ever imagined the cost of experimentation for these newly designed aerofoil?
Depending upon the experiments the cost may range from few lacs to several crores. What if these experiments fail? Absolutely the first thing is direct loss to that amount. Secondly, there is an advantage to the competitor because he had already saved that money.
So the question is, Is there something which we can do to save this money?
The answer is yes and this lies in the description above. As mentioned above, we are closely approximating the physical phenomenon using mathematical equations but we are facing difficulties in solving them. So, it would be better if we generate the equations and let some machine compute it. This is the mother thought of invention and development of simulation science. This science is still under development phase but the progress is really astonishing. Now we can simulate from the stresses to the complicated flow using computers.
One of the simulation science deals extensively with the fluid flow and is named as ‘Computational Fluid Dynamics’ globally adopted as abbreviation of CFD. By seeing the beautiful pictures obtained in the results, some people also call it as Colourful Fluid Dynamics .
For me, it is a combination of both art and science. I can surely justify it for you. Most of the people reading this blog are comfortable with MS Word, but how many of us are a good writer?
There is a mechanic in a garage who knows how to use a drill and an engineer is also there. Both know how to drill but does the mechanic knows where to drill? Similarly, one may say a mathematician and a computer graduate can together solve an equation using computer. The mathematician will write the algorithm to solve while the computer science fellow will write the program for the same. But can we say confidently that they understood the physics of the problem? The results they obtained using CFD tools are correct? Do the results make sense?
In fact, if there is improper boundary conditions, or improper grid or even improper method to solve the equations, the results obtained will be erroneous. I will be similar to using a drill for tightening a screw.
There are a lot of solvers commercially available like Ansys Fluent, CFX, OpenFOAM StarCCM etc. But we need a proper understanding of the subject before proceeding toward the subject. Here in Bangalore I have seen a lot of training institutes claiming to train in CFD. Personally I have tested few of them but unfortunately there is only one training institute which gives proper knowledge and guidance in CFD. It is not like I am doing advertising for them but seriously the way MyCADCFD () handles the concepts and apply it the software tools is really astonishing and interesting. They said they are having two branches in Pune and Bangalore. They also do online training.
In further blogs, we will talk about some interesting cases and results of CFD.[/LEFT][/RIGHT][/RIGHT]
Total Comments 1
Comments

Hello I am a beginner to CFD and currently a student. I was wondering where a good place to start to learn CFD is such as websites, books, and tutorials, so that i can learn to code for simple flows over an object.
Posted November 30, 2016 at 22:13 by stefcfd