1

Calibration using GENOUD optimization algorithm in PAT

Hi,

I trying to calibrate a model using the utility data. I read in a paper on the subject which is titled “ A GRAPHICAL TOOL FOR CLOUD-BASED BUILDING ENERGY SIMULATION” that “to minimize the difference between actual and model energy use. Therefore, we selected the GENOUD optimization algorithm and an objective function that corresponds to CV(RMSE).”

It seems that the GENOUD method is replaced now in PAT with the RGENOUD. I assume these are very much the same. The question is: how to put an objective function that corresponds to CV(RMSE) in PAT. And how does the optimization method work based on this function?

mokhtar's avatar
183
mokhtar
asked 2018-09-04 05:38:49 -0500
__AmirRoth__'s avatar
4.4k
__AmirRoth__
updated 2018-09-04 08:48:25 -0500
edit flag offensive 0 remove flag close merge delete

Comments

add a comment see more comments

2 Answers

3

Once you have one of the the calibration reporting measures described in this post go to the Outputs Tab in PAT. Once you select the calibration measure you should see cvrmse and nmbe outputs each for both gas an electricity. Set the objective column to true for these, and weight them if desired. After this you are ready to go to the run tab to run the analysis.

David Goldwasser's avatar
20.4k
David Goldwasser
answered 2018-09-04 08:45:46 -0500
edit flag offensive 0 remove flag delete link

Comments

Very useful. Thanks a lot.

mokhtar's avatar mokhtar (2018-09-06 01:45:54 -0500) edit
add a comment see more comments
0

to answer your other question, Genoud and RGenoud in PAT are the same. It was just a name change in the GUI but the underlying algorithm here remains the same.

BrianLBall's avatar
2k
BrianLBall
answered 2018-09-04 09:46:32 -0500
edit flag offensive 0 remove flag delete link

Comments

Thank you very much.

mokhtar's avatar mokhtar (2018-09-06 01:46:06 -0500) edit
add a comment see more comments