Learning Forum

[aaron.ryan]

Hi all,

 

For the past number of months I have been working on simulating UV LEDs in Ansys Fluent to determine disinfection rate in a UV LED flow-through reactor. This reactor is used to disinfect drinking water or waste water.

Prior to setting up a full simulation of that reactor, I would like to set up a model for a bench scale version of the reactor (see attached) where a sample is in a petri dish and being constantly stirred. I have data for how well this bench-scale unit works and I would like to simulate it to confirm disinfection rates prior to building a larger model.

 

I’ve managed to create simulations in Fluent and Speos but I’m unable to get the disinfection data to connect. I’ve found and Ansys Healthcare webinar on "UV Fluid Disinfection" presented by Tomasz Stelmach that walks through the process using Speos 2021 R1 with the Optics Multiphysics tool but I haven’t been able to find such a tool, nor have I found a way to generate an interpolation file. I’ve also read and walked through the article on Ansys Optics titled "UV-C surface disinfection of a medial mask", which is what brings me to this post: I’m looking for assistance in setting up a simulation in Speos and exporting the disinfection data to Fluent to determine the disinfection rate in a moving volume of water.

 

I hope I’ve provided enough detail. Any help is greatly appreciated.

[Rob]

I need to be a little careful as I'm not permitted to share nonpublic data and have modelled these systems before. 

Fortunately, knowing the answer makes finding the question to search for easier....  Try Google "uv disinfection with fluent", and you'll see a ScienceDirect paper (I'm also not permitted to share nonAnsys links): I've briefly read the public sections, and don't know anything about the authors. Speos will give you the light/UV intensity in the domain: the fluid media and solid surfaces will effect this, and give you a resulting UV intensity. That intensity can be passed to Fluent, have a look at User Defined Memory and/or the DO model. In Fluent you then need to figure out how long something is exposed to what intensity, DPM Scalars are a good thing to read up on. 

[aaron.ryan]

Thanks, Rob. I appreciate the suggestions.

I'll look into them and make another post if I run into any further issues.