CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   Main CFD Forum (https://www.cfd-online.com/Forums/main/)
-   -   prescribed inflow conditions (https://www.cfd-online.com/Forums/main/501-prescribed-inflow-conditions.html)

 Thorsten January 29, 1999 05:35

prescribed inflow conditions

Hello everybody!

Who could recommend on the prescription of inflow boundary and/or initial conditions for 3D turbulent open-channel flow? Who has made experiences with different conditions? All suggestions in any form (formulas, references etc.) are welcome. Thank you very much in advance for any effort.

Cheers, Thorsten

 John C. Chien January 29, 1999 14:47

Re: prescribed inflow conditions

I think, you are talking about the non-uniform inlet condition. There are several approaches: (1). extend the inlet location in the upstream direction, and specify a condition there. ( you hope that when the flow arrives at the real inlet station it will behave in the way you like.) (2). specify the boundary layer thickness, and prescribe the profiels analytically using available data. ( you can control the non-uniform profiles.) ,(3). compute the section of the flow upstream of the inlet separately. ( this way, you have some ideas about the flow at the inlet station.), (4). measure the flow properties at the inlet station and use it in the computation. (if the flow you are going to calculate does not exist , then this method is useless.). I like your question, because this will tell who is the CFD mechanics and who is the CFD engineer. The proper selection of the computational domain and the associated boundary conditions require the real understanding of the problem itself.

 Thorsten January 29, 1999 16:03

Re: prescribed inflow conditions

Dear John,

Thanks for your response. Yes you are right possibility No.1 is difficult. I will try a combination of 1. , 2. and 3. first using the logarithmic velocity profile law and having Ushear and Umean computed previously with a 1D calculation. Unfortunately No.4 is not always feasible and very time consuming in addition. I hope this might be worth a try. So thanks again and have a nice day. Best wishes Thorsten

 All times are GMT -4. The time now is 15:21.