Frequently used software SIMDEUM now even more accessible

Generating drinking water demand patterns open source in Python

The SIMDEUM software can generate drinking water demand patterns down to the level of individual taps and per second. To increase accessibility to this software, SIMDEUM is now also available as open source software and in the Python programming language.

The new Python version of SIMDEUM (pysimdeum) has several advantages. First, more and more people are working with Python and less with Matlab®. Second, Python is cheaper, especially when simulations are done on a large scale and multiprocessing is required. Third, the code has now been converted to fully object-oriented coding, making further extension easier. In recent years, dozens of PhD and MSc students have successfully used the Matlab® version of SIMDEUM. The creators hope that many more users will follow and that with this version, they will need less support from KWR researchers.

Using pysimdeum

With pysimdeum, thousands of unique (stochastic) drinking water demand patterns can already be easily generated at various scales: from seconds to minutes, quarters or hours and from tap point to water meter, street or an apartment building with dozens of homes. A simple link is possible with EPANET, the software for modelling drinking water distribution systems, and with other hydraulic calculation programs.


Further additions

At present, pysimdeum does not yet have completely the same functionality as the original Matlab version. The KWR researchers hope that soon the modelling of non-domestic drinking water demand patterns, wastewater patterns, and patterns for energy consumption for hot tap water will also be implemented in this version. Because pySIMDEUM is open source, they hope that a growing community of users will also add these and other (new) functionalities.


Open source license

Pysimdeum will be made available for use (via Python Package index), based on Python 3.8+, with several standard Python packages and for further development (via Github) to anyone who commits to the open source license. Documentation will be posted on gRead the Docs . A comprehensive review paper is expected to be published by the end of this year, from which information on the input parameters for SIMDEUM can be distilled for many regions of the world. The input parameters for the Netherlands are already well known (and validated) and available as standard. Pysimdeum was created in a collaboration between KWR (Mirjam Blokker, Bram Hillebrand) and David Steffelbauer (first TU Delft, now Kompetenzentrum Wasser Berlin – KWB).

Result of pysimdeum: water tapping pattern of a building with 20 apartments per second (light blue,) and moving average per 2 hours (black)