|
[Sponsors] |
October 3, 2017, 05:10 |
How to time average my results?
|
#1 |
Member
Roy
Join Date: Sep 2017
Posts: 80
Rep Power: 9 |
Hello everybody
I want to time average my results in CFX-Pre. I found some topics about it in this site and read them but it couldn't help me very well. I also couldn't find time averaging in tutorial. Because in one of the opinions I read I should define my (start averaging time- end averaging time) but I can't find such thing in any place. the only thing I can see is the attached picture. my question is that: Is ((arithmetic average)) time average?If it is so, why it doesn't have (start averaging time- end averaging time)? WHY IS IT DEFINED ON ITERATIONS? By the way my time is 150 seconds and I want to average over the last 130 seconds. My time step is 0.05. If arithmetic time averaging is the place I should use for averaging my results, What should I put for start iteration list and stop iteration list? |
|
October 3, 2017, 09:30 |
|
#2 |
Senior Member
Erik
Join Date: Feb 2011
Location: Earth (Land portion)
Posts: 1,188
Rep Power: 23 |
Arithmetic average sounds like it just averages the numbers for all iterations with no time step weighting. You have constant time steps, so it is equal to a time average in your case.
Simulation Time/Time step = Iteration number Starting Iteration: 20 sec / 0.05 sec = 400 Ending Iteration: 150 sec / 0.05 sec = 3000 The documentation will have more details I'm sure. I'm just guessing here. |
|
October 3, 2017, 11:49 |
|
#3 | |
Member
Roy
Join Date: Sep 2017
Posts: 80
Rep Power: 9 |
Quote:
thank you so much for your answer. So what I understood from your opinion is that: using arithmetic average is appropriate for time averaging in my case because the time step is constant and does not change in solution procedure. right? I'm curious What would happen if I have different time steps in my solution? By the way, If I want to use my steady state solution as my initial guess. Where should I define it? I thought that I can ignore using initial guess in CFX-Pre and in CFX-Solver in define run(attached picture) use my steady state results as initial guess. is that right? |
||
October 3, 2017, 17:33 |
|
#4 |
Senior Member
Erik
Join Date: Feb 2011
Location: Earth (Land portion)
Posts: 1,188
Rep Power: 23 |
Correct on all accounts as far as I understand.
If you had different time steps I'm guessing it would still just average the numbers without weighting them by time duration. I would use "time integral" instead if you have different time steps. See CFX-Pre user guide for all the information on these options and settings. It has all you would want to know and more. |
|
October 3, 2017, 17:37 |
|
#5 |
Senior Member
Join Date: Jun 2009
Posts: 1,880
Rep Power: 33 |
Time averaged values are time step weighted
Time_average = Sum ( phi_1 * dt_1 + phi_2 * dt_2 + ....) / Sum(dt_1 + dt_2 + ...) = Sum (phi_1 * dt_1 + phi_2 * dt_2 + ....) / Total Time It is not an arithmetic average. |
|
October 4, 2017, 18:05 |
|
#6 | |
Member
Roy
Join Date: Sep 2017
Posts: 80
Rep Power: 9 |
Quote:
thanks for your answer; but do you know how and where should I define it in CFX? is there any place I can do it? |
||
October 4, 2017, 18:08 |
|
#7 | |
Member
Roy
Join Date: Sep 2017
Posts: 80
Rep Power: 9 |
Quote:
thanks for your answer. I'm using arithmetic average now. I wish it will be the expected definition I searched. |
||
Tags |
time average in cfx |
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
[General] Extracting ParaView Data into Python Arrays | Jeffzda | ParaView | 30 | November 6, 2023 22:00 |
Floating point exception error | lpz_michele | OpenFOAM Running, Solving & CFD | 53 | October 19, 2015 03:50 |
Moving mesh | Niklas Wikstrom (Wikstrom) | OpenFOAM Running, Solving & CFD | 122 | June 15, 2014 07:20 |
Micro Scale Pore, icoFoam | gooya_kabir | OpenFOAM Running, Solving & CFD | 2 | November 2, 2013 14:58 |
Upgraded from Karmic Koala 9.10 to Lucid Lynx10.04.3 | bookie56 | OpenFOAM Installation | 8 | August 13, 2011 05:03 |