Script:Unsupervised k-medoids typical days

From HUES Platform Wiki

To edit the entries in this page, select "Edit with form" in the "Page tools" menu at the top of the page.


General information

Description The typical day method used in Dominguez-Munoz et al. (2011) are implemented in a MATLAB code. The k-medoids clustering method is employed selecting k typical days able to represent the full year demand pro file. 3 peaks days for heating, cooling and electricity demand pro files are added. The k number of days is selected by the user. The unsupervised typical days function select the number k of typical days which best represents the full year demand profile. This is done based on a minimization of error between the load duration curves for heating, cooling and electricity and the Davies-Bouldin index representing the intra and inter clusters relation (Davies and Bouldin (1979)). The user select the number k of typical days depending on the pareto front and the computation time.
Download URL https://bitbucket.org/empa_lbst/unsupervised_kmedoids_typical_days/get/HEAD.zip
Authors Julien Marquant, Georgios Mavromatidis
Required software - MATLAB 2014 or upward
Related publications [[Related publications::Selection of typical demand days for CHP optimization, Fernando Domínguez-Muñoz et all, 2011 [1]]]
License type MIT License
Tags typical days


Documentation

Documentation may be uploaded, linked or directly entered into the wiki.

Documentation URL(s):
Documentation file: File:Guidelines Non-naive typ days.pdf
Documentation page: