Learning Forum

[aCVP]

Hi all - I have a question about ANSYS Discovery Live.

I was told that Discovery Live is not good for blood flow simulations. However, I found an article of a Discovery Live competition where someone looked at vorticity/thrombus formation. Here is the article:

https://www.ansys-blog.com/winner-discovery-live-design-competition-may-save-lives/

I would love to be able to use Discovery Live in our blood flow simulations, but I would like to know the accuracy in comparison to Fluent.

Any thoughts would be appreciated!

Thanks!

[aCVP]

To provide more information: I will be looking at blood flow within ventricles. 

[jmccasli]

Hello aginn,

 

Discovery Live is meant as a preliminary design tool in the very early stages of product design and development. It is great for detecting trends when making large-scale changes to a model, but it is not meant studying small scale details of a flow field. Here's a good example I heard from the Discovery Live product manager, when using DL for a multi-element airfoil study with a Formula 1 team: DL will do a good job at capturing whether or not total system drag goes up/down when moving the airfoils relative to one another. But it will not do a good job of capturing the detailed velocity profile or shear in narrow passages between airfoils.

 

Using this same line of thinking on your blood flow within ventricles, it is safe to say that DL can give you useful information about global velocity features and overall system pressure when changing X or Y inputs. But it would not be good to rely on the actual values of, e.g., ventricle wall shear for a certain design.

 

In short, DL is useful for making early design decisions. The key thing to remember here is that DL is meant to be used in complimentary fashion with Fluent, not as a replacement for Fluent. E.g., you could run 100 designs in DL very quickly, then the top 5 candidates could be run in Fluent for more detailed analysis.

 

I hope that helps.

 

Regards,

Jeremy

[Rob]

DS also doesn't have the nonNewonian viscosity that you'll need as blood isn't a simple liquid (it's actually a slurry). 

[aCVP]

Can anyone direct me to documentation about the numerical methods that DL uses? I would love to learn more about it. 

[Rob]

If it's not in the Help guide there's nothing we can add here: we're very limited what we can discuss on a public forum.