Question-and-Answer Resource for the Building Energy Modeling Community
Get started with the Help page
Ask Your Question

Sequential load distribution scheme seems not working

asked 2017-11-12 06:45:00 -0500

katsuya.obara's avatar

updated 2017-11-20 08:39:24 -0500

I am modelling with multiple water cooled chiller in energyplus.
In idf file, I specified SequentialLoad for Chilled Water Distribution Scheme. However, looking at the result, it seems that multiple chiller is operating before 1 chiller get to full load as shown below graph.(x:part load ratio/y:COP)
image description
Does anyone know why this happen?

Timeseries stack plot and stack plot by ascending total cooling rate is added following @aparker instruction.
Looking at the 2nd graph, I can see the fluctuation when the consequent chiller start to be operated. Is this usual? or something is causing this fluctuation?

image description

image description

edit retag flag offensive close merge delete



It generally seems to be working as intended; not sure if the black in the graph is ChillerB3 or if it the top of each series has a black border. Also, I forgot to mention you should be using "Detailed" as the output frequency. The HVAC systems operate on a sub-hourly basis, so if you use "Hourly" you can see some confusing fluctuations which are just the effects of the hourly averaging.

aparker's avatar aparker  ( 2017-11-20 15:58:04 -0500 )edit

@aparker Thank you for your comment. Do you know how I can interpret the fact that each chiller doesn't operate to maximum part load ratio(100%) even though I specify Sequential Load for Chilled Water Distribution Scheme?

katsuya.obara's avatar katsuya.obara  ( 2018-06-27 21:32:02 -0500 )edit

2 Answers

Sort by ยป oldest newest most voted

answered 2017-11-18 18:13:34 -0500

updated 2017-11-25 11:44:12 -0500

@TomB is right, and the chart he suggested should be your starting point:

  1. Report the timeseries cooling output for every chiller
  2. Graph them all on the same plot

It will probably look something like this:

image description

Graphed this way, I find it challenging to see if the overall staging scheme is working correctly without a lot of zooming in on the details of each day unless the staging scheme is very simple. Instead, I like to use those same outputs to create a different graph to understand how the chillers (or boilers, or other plant equipment) are staging:

  1. Sum up the cooling output for each chiller
  2. Sort the timeseries from smallest to largest total cooling output
  3. Plot a stacked timeseries of the sorted results.

Three chillers running sequentially should look like the image below.

image description

image description

The graph below shows actual simulation output for a more complex staging scheme. There are 3 chillers and 4 different capacity ranges. Only certain chillers are used inside each capacity range, and within each range, the chillers stage 'optimally'. As you can see, this chart makes it very easy to confirm that the model is working as intended.

image description

A nice bonus of this graph is that you can easily see how much of the year is spent in each operating range.

image description


First, a clarification: images 2 and 3 are just drawings in powerpoint, not actual simulation data. Images 1, 4, and 5 are actual simulation data.

Second, here are the key objects/inputs for getting staging right:

  1. Chiller:Electric:EIR Sizing Factor. The sizing factor field controls the sizing of each piece of equipment on the plant loop, and the default is 1 (100%). If you have 3 chillers, all with a sizing factor of 1, then each chiller will be sized to meet 100% of the load of the plant loop. This means you'll end up with 300% of the required capacity, and only 1 chiller will ever run. If you want to get 3 equally sized chillers, which, combined, would meet the total cooling load, you would set the sizing factor to 0.33 in each chiller. Also, the sizes don't have to be equal, you could have a 0.2, a 0.3, and a 0.5 or some other combination.

  2. PlantEquipmentOperation:CoolingLoad. This object determines which pieces of equipment are allowed to operate for a given range of capacities. By default, there is only 1 range (0+ W), and every piece of equipment is allowed to operate. The 4th image in this post shows the results of a more detailed staging scheme:

    • 0-25 tons: Small Chiller
    • 25-50 tons: Medium Chiller
    • 50-75 tons: Small Chiller, Medium Chiller
    • 75-100 tons: Medium Chiller, Big Chiller
    • 100+ tons: Small Chiller, Big Chiller

    It is important to understand that allowed to operate just means that the equipment may be used, not that it will be used. Determining which pieces of equipment do operate is the role of the next field:

  3. PlantLoop Load Distribution Scheme. This ...

edit flag offensive delete link more


@aparker your first two charts have very sharp boundaries, is this simulated data, or just an example of a chart? Could you expand on your answer to suggest what factors you think influence a succesfuly staging strategy? i.e any tips and tricks you've learnt that lead to a nice tidy result like your fifth chart

TomB's avatar TomB  ( 2017-11-22 16:20:12 -0500 )edit

@TomB edited per your comments.

aparker's avatar aparker  ( 2017-11-25 11:44:37 -0500 )edit

answered 2017-11-16 21:08:15 -0500

TomB's avatar

Your question is too broad, and there is not enough information provided, to provide a concise answer. There are a number of things you can check - What capacity does each chiller have - have you autosized the chillers using a 'Sizing Factor' set at the default of 1? - The chart presented is not a good choice for the question you've posed - to definitively say that more chillers are operating than you expect at any given time step, you need to plot the time series on the x axis. (it is a nice looking chart though !) - The chillers may not get to 'full load' at every time step. the capacity of the chiller will vary as the condenser water and chilled water temperatures vary, so they may be operating at max. available capacity - look into the CW temperatures and CHW temperatures at unexpected operating PLR/COPs

edit flag offensive delete link more

Your Answer

Please start posting anonymously - your entry will be published after you log in or create a new account.

Add Answer

Training Workshops

Question Tools



Asked: 2017-11-12 06:45:00 -0500

Seen: 36,794 times

Last updated: Nov 25 '17