A Clustering Based Methodology for Natural Gas Demand Analysis

Natural gas is considered as a crucial power source both for electrical and thermal loads. Natural gas markets are continually expanding, a fact that raises the need of formulating new energy services to the end-users. The aim of the present paper is to develop a two-stage clustering process that le...

Full description

Saved in:
Bibliographic Details
Main Authors: Panapakidis, Ioannis P., Bouhouras, Aggelos S., Christoforidis, Georgios C., Parisses, Constaninos
Format: Conference Proceeding
Language:English
Subjects:
clustering
Clustering algorithms
Clustering methods
Demand side management
Machine learning
Natural gas
natural gas demand
Power markets
thermal load profiling
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Natural gas is considered as a crucial power source both for electrical and thermal loads. Natural gas markets are continually expanding, a fact that raises the need of formulating new energy services to the end-users. The aim of the present paper is to develop a two-stage clustering process that leads to the formation of natural gas demand profiles per distribution point. Natural gas demand profiles can be further utilized in a variety of applications such as forecasting, tariff designs and others. The test case refers to the distribution points of the national natural gas system of Greece. The results show that hierarchical algorithms are the suitable clustering solution for the dataset under study.
ISSN:2166-9546
DOI:10.1109/CPE.2019.8862407