Ask Your Question
3

Energyplus: is it possible to increase the variability of power demanded for a chilled water plant?

Hi everyone!

I have this Energyplus model which simulate a hospital building energy consumption. The building has 3 air-cooled chillers with a total capacity of 220 TR which deliver only cooled air. The system has 27 fancoils and 4 pumps without modulation. Comparing the real and simulated plant shows a big deviation: to illustrate, I plotted the measured data and Energyplus output (time axis over X) for a little period (temperure of outdoor air included).

image description

As you might remark, the measured data has a great variability! To put in numbers, the standard deviation for measured data is about 34 kW (whole measurement period), while the simulation stands for 5 kW (almost seven times lower).

I am wondering which setting I can change in my model in order to increase the power demand variability for the chilled water plant. Some attempts that I already tried without success:

 - Change Chiller COP;  - Change Chiller Flow Mode from NotModulated to LeavingSetpointModulated;  - Change constant speed pump to variable speed;  - Change constant volume to variable volume.

Someone can help me in this issue?

ppfsilva's avatar
55
ppfsilva
asked 2017-10-25 16:22:10 -0500, updated 2017-10-27 13:13:15 -0500
edit flag offensive 0 remove flag close merge delete

Comments

add a comment see more comments

1 Answer

3

I am assuming that the X-axis for the graph is time. If so, based on the graph - it appears that the real-life system has some schedule attached to it as the data trend is repetitive. The changes that you made on the plant side make sense from the supply side. Since the chiller operation follows the demand in your energy model, I would now start looking at the schedules that impact the cooling from the demand side: 1. Schedules for internal loads (particularly occupancy); 2. Thermostat schedules; 3. Schedules for fan coil unit fans;

rsunnam's avatar
960
rsunnam
answered 2017-10-26 05:32:21 -0500
edit flag offensive 0 remove flag delete link
more

Comments

Thanks for your answer. For sure your recommendations returned an increase in variability of simulated chiller, unfortunately it was not enough! I'm getting to the conclusion that I really can't simulate this plant consumption.

ppfsilva's avatar ppfsilva (2017-10-27 16:07:02 -0500) edit
more
  • delete

Glad that was a good start. It is a tough problem to make sure that the energy model consumption exactly matches the actual consumption. If you have access to the BAS sensor data, I would suggested that you populate all your mech system schedules from those. Also, ensure that your internal gains' schedules are as accurately modelled (may be interviews with facilities manager?). For similar projects, I often found the system operation was usually not as expected based on BAS data. So, I usually visualize BAS data against the energy data for such comparisons. It is usually an iterative process.

rsunnam's avatar rsunnam (2017-10-27 17:09:52 -0500) edit
more
  • delete
add a comment see more comments