Movement of residence time geometries model

Shinichiro_Oshiro

I would like to use the plugin residence time model to analyze the residence time in a specific area of ​​a translating geometry.

Is it possible to move the analysis area to match the movement of the geometry?

 

Best regards,

Oshiro

RWood

Hi Oshiro,

Unfortunately with the current implementation this isn't possible - at least not without a lot of brute-force hacking of files (which I'm not about to recommend). I don't know how complicated your simulation is but the only way I can think of doing this is moving the rest of the model to keep the volume used for the residence time calculation stationary. This might involve rotating the gravity vector though, depending on how complicated your model is. 

This sounds like it could be relatively straightforward to implement, so I'll discuss internally about the possibility of adding this. However, if we did add this it's unlikely to be anytime soon.

If there are updates I'll let you know.

Cheers,

Richard

Shinichiro_Oshiro

Hi Richard,

 

I understand that it is currently not possible.

I look forward to your future developments.

 

Best regards,

Oshiro

Ignacio Diez Alonso_21259

Hi Oshiro,

I wanted to add to what my colleague has mentioned. That even though the API model may not be able to do this. Since the release of EDEM 2023.0 it is possible to look at and export Residence time in Geometries from the Analyst. This can be done with a selection, with the added benefit that selections can be linked to other Geometries to follow their motion. Hope that this helps.

Kind regards

Ignacio

RWood

Ignacio Diez Alonso_21259 said:

Hi Oshiro,

I wanted to add to what my colleague has mentioned. That even though the API model may not be able to do this. Since the release of EDEM 2023.0 it is possible to look at and export Residence time in Geometries from the Analyst. This can be done with a selection, with the added benefit that selections can be linked to other Geometries to follow their motion. Hope that this helps.

Kind regards

Ignacio

For clarity though, this is the residence time spent in the simulation, not the geometry. Using this approach will show the correct particles, but the residence time value will not be the time spent in the geometry itself, but the simulation overall.

Shinichiro_Oshiro

Hello, Ignacio and Richard.

Thank you for your responses.

When the geometries created by the "Selection" and Residence time APIs are the same in terms of shape and position, will the residence times outputted by each be different?

 

Best regards,

Oshiro

RWood

Shinichiro Oshiro said:

Hello, Ignacio and Richard.

Thank you for your responses.

When the geometries created by the "Selection" and Residence time APIs are the same in terms of shape and position, will the residence times outputted by each be different?

 

Best regards,

Oshiro

Hi Oshiro,

It depends. The property 'Residence Time' in the Analyst (the one you can analyze with selections) is simply the time the particle has spent in the simultation, no matter where it has been. It's essentially:

current simulation time - particle birth time

The 'Residence Time in Geometries' models calculate their property (Custom Residence Time if I remember correctly) only if the particle has spent a timestep in the bounds of the specified geometry, i.e. the residence time value is incremented by the value of the timestep if this is the case.

If you specify the bounds of the geometry to overlap the entire domain then the two approaches will return the same value, in the general case they will not be equal though.

Hope that makes sense?

Richard

Shinichiro_Oshiro

Richard Wood_20774 said:

Hi Oshiro,

It depends. The property 'Residence Time' in the Analyst (the one you can analyze with selections) is simply the time the particle has spent in the simultation, no matter where it has been. It's essentially:

current simulation time - particle birth time

The 'Residence Time in Geometries' models calculate their property (Custom Residence Time if I remember correctly) only if the particle has spent a timestep in the bounds of the specified geometry, i.e. the residence time value is incremented by the value of the timestep if this is the case.

If you specify the bounds of the geometry to overlap the entire domain then the two approaches will return the same value, in the general case they will not be equal though.

Hope that makes sense?

Richard

Hi Richard,

Thank you. I understood.

I found that the residence time in the Analyst was not enough because I wanted to know the time spent in a specific area.

Best Regards,
Oshiro