Evaluation of spatial clustering methods for regionalisation of hydrogen ecosystems
Spatially resolved modelling of local hydrogen ecosystems can help to identify optimal sizing and locations for plants and infrastructure along the value chain. Spatial clustering to identify the subregions can lead to a better representation of important features compared to administrative units or...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-01-01
|
| Series: | Energy Strategy Reviews |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2211467X24003365 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Spatially resolved modelling of local hydrogen ecosystems can help to identify optimal sizing and locations for plants and infrastructure along the value chain. Spatial clustering to identify the subregions can lead to a better representation of important features compared to administrative units or uniform grids. Several algorithms are available for regionalisation, but an evaluation of their suitability for hydrogen ecosystems or similar applications is missing in the literature. This paper presents a holistic evaluation of different spatial algorithms based on existing and newly developed statistical indicators. Although the best algorithm depends on the focus of the regionalisation process, the method REDCAP proved to be the best overall, especially with higher intra-cluster homogeneity compared to the widely used k-means algorithm. The developed indicators and their evaluation regarding different objectives are seen to be transferable to other clustering and regionalisation applications, like energy system analysis or general supply chain analysis. |
|---|---|
| ISSN: | 2211-467X |