Significantly higher chiller energy use in v25.1 for primary-secondary common pipe and branch pumps

[mjwitte]

Issue overview

From helpdesk ticket 16933

User: Before upgrading to EnergyPlus 25.1 this configuration worked more or less as expected. But, after the upgrade, the chiller cooling energy went up significantly.
MJW: So, the big change happened between 24.2 and 25.1. I see rather different patterns of chiller operation in 25.1 vs 24.2. I wonder if it's related to this change:
#10793
MJW Note: Made a build of 24.2 just after PR10793 (really pre-25.1) and the bad chiller results are there, same as full 25.1.
Made a build of 24.2 just before PR10793 (really pre-25.1) and the results are the same as 24.2 release.
Changed the output reporting to detailed with 24.2 release and checked how many timesteps had a chiller evap rate > 0 with chiller electricity rate = 0, and there were none.
So, PR10793 broke this file for no good reason.
User:
I found that in the InputOutputReference that E+ does not support more than one pumps in the primary loop when using "CommonPipe".

Also, I found that the simulation results have never been accurate, but in E+25.1 the issue is magnified.

I calculated the heat transfer rate at the cooling coil level, and I found that whenever multiple pumps are used the heat transferred to water (calculated using coil water flow rate and delta T) is higher than the cooling coil cooling rate. (see LoadCalc.xlsx in DefectFiles).

Switching to a single branch pump does make the results look reasonable.

Possible workaround:

1. Use a single branch pump or loop pump on the primary side.
2. Drop the common pipe option and replace it with two full plant loops using HeatExchanger:FluidToFluid to connect them.

Operating System (Multiple choices)

Any

Operating System Version

any

Version of EnergyPlus

25.1.0 (also 25.2RC1)

Unmethours link or helpdesk ticket number

Helpdesk ticket 16933

Defect file

EnergyPlusDevSupport\DefectFiles, Sacramento Executive airport weather

[mjwitte]

v24.2

v25.1

v25.2RC1 (similar to 25.1)

[mjwitte]

Additional info from user:
We decided to move away from the CommonPipe approach and started using the HeatExchanger approach. The HeatExchanger approach does resolve our issues. Maybe EnergyPlus should just disable the CommonPipe approach.