Compute temperature for xflowing injectors as weighted average #5799
Conversation
…ids entering the well.
|
jenkins build this please |
|
I am not familiar with the energy conservation equations. But ideally, should we have a primary variable to monitor the temperatures in the wellbore to handle the crossflow through heat conservation? |
Maybe @hnil and/or @totto82 can comment on that. I'm just trying to get slightly more representative output with the current approach.. |
|
If we are doing this only for output. I think a post processing like @vkip is suggestion is ok. But if we want to use this computed temperature for the energy equation I think we need to add a variable and solve for it as @GitPaean suggests. Maybe we should start investigating this for MSW? For STW we will only have one temperature for the whole well. For cross-flowing wells that may be a too coarse approximation. |
| const Scalar vol_heatcap = cellDensity/cellBinv * cellInternalEnergy/cellTemperature; | ||
| const Scalar heatcap_rate_prod = -std::min(perfPhaseRate, 0.0)*vol_heatcap; | ||
| const Scalar heatcap_rate_inj = perfPhaseRate*vol_heatcap; | ||
| total_weight_prod += heatcap_rate_prod; |
There was a problem hiding this comment.
Why is this (line 2907) not std::max(perfPhaseRate, 0.0)? (to get the inflow?)
There was a problem hiding this comment.
Because I want to end up with the net injected rate for the well (weight for the injection temperature).
For what it's worth, I would support both, assuming adding a variable for STW would be an improvement. (But separate from this PR..) |
|
jenkins build this please |
Weighted average of fluids entering the well from both surface and reservoir.