[pano]

Hi all,,

I want to use CFD in my MEng project for the 2 following cases and I would like your opinion / advise on that.
1. The project is the design of a compressed natural gas carrier ship to transfer natural gas (NG), it is an alternative means to transfer NG as opposed to LNG. Anyways, more to the point, the compression takes place on board, that means the NG is initially loaded at relatively low pressures (about 50 bar) and then it has to be cooled and compressed until it reaches -30C and 250bar. The compression and cooling takes place in successive boxes and then the NG is stored in pressure vessels. I know it is possible to model all these thermodynamics in MATLAB for example using Simulink pretty much like modelling an ICE. However, I think I can use CFD for these heat transfer calculations. Then I hope I will be able to determine the time required to compress and chill a certain amount of gas and also give me an insight to design properly these compression rooms so that I achieve the maximum efficiency of the loading process. Simirarly the unloading process is pretty muh the opposite of the loading process.

2. I also intend to model heat loss of natural gas stored under high pressure in these composite pressure vessels (long tubular tanks). The NG inside the PVs is cooled at -30C and compressed at 250 bar (but still remains in its gaseous state of course, unlike LNG). The gas is cooled in the loading process (see above) and then stored in the insulated PVs until the end of the journey. The aim is to investigate how the temperature inside these insulated tanks increases with the time resulting in pressure increase as well. The point is that due to these extremely high pressures, the PVs are not able to withstand much more than 250 bar (even though they are indeed designed with high safety factor anyways), so I need to know how exactly the heat transfer takes place. Of course this is all possible to some extent with basic insulation formulae and heat transfer coefficients and everything, but do you think I can get reliable results using Fluent ?

Just a note: I am fairly familiar with Fluent and Gambit as I did my thesis last year using these programmes. In fact, I didnt consider any heat transfer in that project as it was purely hydrodynamic study, but I consider myself familiar with the software.

What do you guys think?? Any idea / hint / advise is very welcome. Please ask me to give more details of my study if you think I am not describing it accurately enough here.

Cheers, Pano