Solving Navier-Stokes in DART language
 

Hello All,
I want to solve the Navier-Stokes equations for a steady state, incompressible natural convection problem by coding in DART language. DART is a web programming language developed by Google. I wish to simulate natural convection in a room, heated by a thermostat in a house. My question is: If I want to do the calculations on a cloud (like the Amazon EC2), How much time will it take for convergence? I'm developing this for visualisation purposes only; for children to understand how natural convection looks, so accuracy is not a constraint. This is going to be a part of an app for android.
The problem statement: When a child sets a value on the thermostat, an animation should pop-up which shows the air currents and approx. temperature isotherms in the room.
Amy thoughts/ questions are welcome.

Thanks in advance!

A straightforward solution to your problem would seem to be to use an existing CFD program to generate a set of animations covering the range of the thermostat. Then get your app to simply show the particular animation corresponding to the current setting of the thermostat.

Thanks for your reply. There are other factors that will be considered during the simulation like ambient temperature (there is a window in the house, through which more heat transfer will happen), and different building materials. I was thinking of generating data for some temperature values in the thermostat's range and then use some interpolating algorithm to plot data for the intermediate values. Can I do this for animations?