Authors' implementation of “Segmentation of Low Voltage Consumers for Designing Individualized Pricing Policies” (2017), 14th International Conference on the European Energy Market (EEM).
Paper: https://ieeexplore.ieee.org/document/7981862/
In this paper, clustering algorithms that are considered as cutting-edge technology for Low Voltage consumer segmentation (such as K-means and its variants, the Fuzzy C-means algorithm, the Self-Organized Maps and algorithm Hierarchical Clustering) are investigated. Experiments are associated with the selection of optimum parameters of clustering algorithms, depending on the input data and the selection of the optimal algorithm in each case, in order to highlight the main consumption patterns in three different data sets originated from pilots of various European Research Programs. The clustering results are evaluated using various evaluation metrics that help form useful conclusions about the coherence and consistency of the consumer groups.
In order to show the practical application of the consumer groups that are created as a result of the above analysis, we designed appropriate pricing policies for each individual group. The experimental results show that we can achieve significant reductions in both the consumption peak hours and the consumer's electrical billing costs, proving the correctness and usefulness of our approach to real scenarios.
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License. To view a copy of this license, visit https://creativecommons.org/licenses/by-nc-sa/4.0/.