Soft Dynamic Time Warping (SDTW) Clustering

Source code and data from "Soft Dynamic Time Warping for Clustering Drinking Water Demand Patterns from Smart Water Meters" paper, developed in Steffelbauer et al. (under review)

This repository contains the code of the Soft Dynamic Time Warping Clustering Approach for user segmentation and high-level information extraction from daily water demand patterns. The repository also contains the data used in Steffelbauer et al. (under review) to test the methods.

Requirements:

• Python - The code in this repository has been developed and tested on Python 3.7.1

The following additional Python packages:

• numpy.
• pandas.
• scipy.
• sklearn.
• tslearn.
• seaborn.
• matplotlib.

Citation: - Steffelbauer, D.B., Blokker, E.J.M., Buchberger, S.G., Knobbe, A., & Abraham E. (under review). Soft Dynamic Time Warping for Clustering Drinking Water Demand Patterns from Smart Water Meters. Water Resources Research, under review

Authors

Code: David B. Steffelbauer - Water Management Department | TU Delft - Faculty of Civil Engineering and Geosciences

Paper: Steffelbauer, D.B., Blokker, E.J.M., Buchberger, S.G., Knobbe, A., & Abraham E.

References

• Steffelbauer, D.B., Blokker, E.J.M., Buchberger, S.G., Knobbe, A., & Abraham E. (under review). Soft Dynamic Time Warping for Clustering Drinking Water Demand Patterns from Smart Water Meters. Water Resources Research, under review