Averaging distributions

[filippov1]

Is it possible in Visit to plot distributions of a flow parameters (such as pressure or temperature) averaged over a certain range of iterations or a period of time?
If yes, could you please send me a link to documentation?

Thank you!

[markcmiller86]

If instead of seeing an instantaneous pressure variable, you want to see the pressure variable averaged over some set of timesteps, you can use the Time iteration expressions.

[filippov1]

Hi Mark, Thank you very much for the answer! I am positive that I just need to do it right once. So far, I had errors. My test expression is average_over_time(P,2601000, 2676000,1000) (average of pressure over the whole simulation between the iteration 2601000 and iteration 2676000 with the stride of 1000. I have the following warning: The compute engine running on host rzhound205.llnl.gov issued the following warning: avtAverageOverTimeExpression: Clamping end time to number of available timesteps. The picture looks strange, with extremely high values: [Chart, pie chart Description automatically generated] What did I do wrong? Thank you! Best regards, Andrey From: Mark C. Miller ***@***.***> Sent: Sunday, June 30, 2024 5:15 PM To: visit-dav/visit ***@***.***> Cc: Filippov, Andrey ***@***.***>; Author ***@***.***> Subject: Re: [visit-dav/visit] Averaging distributions (Discussion #19628) If instead of seeing an instantaneous pressure variable, you want to see the pressure variable averaged over some set of timesteps, you can use the Time iteration expressions<https://urldefense.us/v3/__https:/visit-sphinx-github-user-manual.readthedocs.io/en/develop/using_visit/Quantitative/Expressions.html*time-iteration-expressions__;Iw!!G2kpM7uM-TzIFchu!2hZIe1aWpfUvoU0jdsgS6XSPEdh1ekxqZ4Vq2SdT979Uz79cgbJfuLuLJAN8U-730YwYXLE1eVa2bimXTe5aqa8LtFUT$>. — Reply to this email directly, view it on GitHub<https://urldefense.us/v3/__https:/github.com/visit-dav/visit/discussions/19628*discussioncomment-9919866__;Iw!!G2kpM7uM-TzIFchu!2hZIe1aWpfUvoU0jdsgS6XSPEdh1ekxqZ4Vq2SdT979Uz79cgbJfuLuLJAN8U-730YwYXLE1eVa2bimXTe5aqcQDs0Dq$>, or unsubscribe<https://urldefense.us/v3/__https:/github.com/notifications/unsubscribe-auth/AZ5YHZT63UI7E6X5UNRC2CTZKCNRNAVCNFSM6AAAAABKEQ4TZGVHI2DSMVQWIX3LMV43SRDJONRXK43TNFXW4Q3PNVWWK3TUHM4TSMJZHA3DM__;!!G2kpM7uM-TzIFchu!2hZIe1aWpfUvoU0jdsgS6XSPEdh1ekxqZ4Vq2SdT979Uz79cgbJfuLuLJAN8U-730YwYXLE1eVa2bimXTe5aqbQlE3fs$>. You are receiving this because you authored the thread.Message ID: ***@***.******@***.***>>

[markcmiller86]

@filippov1 the integer arguments you pass for start, stop and stride are state-index terms, not in the data producer's cycle or time terms.

So, for example, if you have a data producer (evaluating some PDE) producing 20 output files every 0.1 seconds (in simulation time) starting at time 1.5 seconds (in simulation time) and ending at 3.5 seconds (in simulation time), and you want to average only the stuff in the second half of the run, you would do average_over_time(P, 10,20,1) and NOT average_over_time(P, 2.5,3.5,0.1).

That is, start, stop and stride are an index over the set of available timestates that VisIt can actually read.

Does that make sense?

[filippov1]

Hi Mark, Thank you very much for explanation! The averaging is working fine now. Best regards, Andrey Filippov Modeling and Simulation Computational Engineering Division Lawrence Livermore National Laboratory Livermore, CA 94550 (925) 422-0568 – office (925) 409-6580 - cell From: Mark C. Miller ***@***.***> Sent: Monday, July 1, 2024 12:01 PM To: visit-dav/visit ***@***.***> Cc: Filippov, Andrey ***@***.***>; Mention ***@***.***> Subject: Re: [visit-dav/visit] Averaging distributions (Discussion #19628) @filippov1<https://urldefense.us/v3/__https:/github.com/filippov1__;!!G2kpM7uM-TzIFchu!1YyXAa_V_sjJf1TwVtf2pQYpgfuYUlF5rY38LhAdGht84IhW4l6HJElPTjLT_Sjdwkc1uQ8m71wymnU-4sexxguTUFtx$> the integer arguments you pass for start, stop and stride are state-index terms, not in the data producer's cycle or time terms. So, for example, if you have a data producer (evaluating some PDE) producing 20 output files every 0.1 seconds (in simulation time) starting at time 1.5 seconds (in simulation time) and ending at 3.5 seconds (in simulation time), and you want to average only the stuff in the second half of the run, you would do average_over_time(P, 10,20,1) and NOT average_over_time(P, 2.5,3.5,0.1). That is, start, stop and stride are an index over the set of available timestates that VisIt can actually read. Does that make sense? — Reply to this email directly, view it on GitHub<https://urldefense.us/v3/__https:/github.com/visit-dav/visit/discussions/19628*discussioncomment-9929098__;Iw!!G2kpM7uM-TzIFchu!1YyXAa_V_sjJf1TwVtf2pQYpgfuYUlF5rY38LhAdGht84IhW4l6HJElPTjLT_Sjdwkc1uQ8m71wymnU-4sexxldRrnas$>, or unsubscribe<https://urldefense.us/v3/__https:/github.com/notifications/unsubscribe-auth/AZ5YHZWONJX3IOXNRF4L4T3ZKGROLAVCNFSM6AAAAABKEQ4TZGVHI2DSMVQWIX3LMV43SRDJONRXK43TNFXW4Q3PNVWWK3TUHM4TSMRZGA4TQ__;!!G2kpM7uM-TzIFchu!1YyXAa_V_sjJf1TwVtf2pQYpgfuYUlF5rY38LhAdGht84IhW4l6HJElPTjLT_Sjdwkc1uQ8m71wymnU-4sexxsQY_TNf$>. You are receiving this because you were mentioned.Message ID: ***@***.******@***.***>>