Learning Forum

[Bill Hsu]

Hello, I used the Discrete Phase Model to simulate the trajectories of PM2.5 in indoor space, the volume injection is used to set 4,000 particles uniformly distributed in indoor space, and an air conditioner and an air purifier are used to dilute the particles inside the room. Here's my question:

The number of particles might be different in each cell, some cells have several particles inside, and some of them have no particles inside.  Is it possible to know how many particles in each cell?

 Since I ran the simulation for 30 minutes, the suspended particles are decreased as time passed, I used the summary in Particle track to get the number of trapped, escaped, and suspended particles, but if I want to know how many particles are closed to the ceiling or closed to a specific location, how do I display about that?

The following pic is the suspended particles at 30th min (Flow time).

[Rob]

You can find a particle concentration, which is kg of parcel per m3 cell volume. That may be sufficient. Otherwise it's a little more complicated as DPM positions don't return a cell ID without using a UDF. 

[Bill Hsu]

Thanks for your reply, Rob. Since I use the one-way coupling assumption (Interaction with continuous phase in DPM didn't enable), the DPM concentration cannot be found in every dropdown list. Is there any other way to find the particle concentration?

[Rob]

Ah, if there's no interaction then the tracks are "over" the mesh. So, the particle locations can be known, but not the cells they're in without coding. Are you running the particles as transient?

[Bill Hsu]

Yes, both of the airflow field and particles are run as transient.

[Bill Hsu]

The particle location you mentioned is the current position for the particle?

[Rob]

Yes, if you write the DPM Summary/Report (I can never remember which) it should return the current positions. 

Or, you can cheat. If you run with two way coupling for a particle update you'll get the concentration metrics. It's an old but common approach to mess with the DPM interactions in a variety of ways to get the solver to do things it's not supposed to. However, as they're not in the manual I'm not allowed to go into detail on the public Forum. 

[Bill Hsu]

Thanks for your suggestions, Rob. One more question: how do I show the room frame like the picture shown below? Since I "draw mesh" in Fluent and only enabled "edge", it showed some random lines on the wall of the room model (shown in the pic I attached to this post), how do I eliminate those random lines and just keep the room frame?

[Rob]

Check feature angle is sensible. But if there are holes in a face you'll always see those. One option for the above is to omit including the ceiling & floor in the mesh object. 

[Bill Hsu]

First of all, thanks for your reply, Rob. But I do not quite understand your suggestion. Since changing the feature angle in my model cannot remove the irregular lines on faces, I should try another approach. I wanna make sure if "One option for the above is to omit including the ceiling & floor in the mesh object. " means that omit the ceiling and floor zone in the "Draw mesh" options. My boundary zones only include inlets, outlets, and walls (The ceiling, front wall, side walls and floor are considered as "wall" in name selection ).  

[Rob]

Yes for the walls. You're after showing a shape using edges. If I want to see the shape of the room I generally just display the walls if I can assume floor & ceiling are flat. 

[Bill Hsu]

Thanks for your reply, Rob. After I only enable the walls in the Draw mesh option, the random lines on faces still exist, so I used the CFD post to display the particles in the room instead of using Fluent. But I still don't know why the random lines show on the model's faces. The following figure shows the particles in the room in CFD post. 

[Rob]

Ah, that's the partition lines - sorry I missed what you meant. It's an option in the Mesh display, but I can't remember when it went into the code.