Round about half of the new H2-Pipelines are planed as repurposed CH4-Pipelines
After Implementation of these new H2-Links the repurposed CH4-Pipelines, which the H2-Links will be replace, has to be deleted
[ ] It is necessary to create a method, which is able to match the new repurposed H2-Links with the according CH4-Links for deleting the relevant CH4-Links
First approach to find matches by comparing the distance of start and end points of H2-Links and CH4-Links is not useful, cause the accuracy of H2-Pipelines is not high enough. For large buffer radii, wrong matches between CH4 and H2 pipelines are found, for small buffer radii, a match is only found for a very small proportion of the repurposed H2 pipelines
Ulf's Appraoch (not tried out yet) :
For each conversion pipeline, find the nearest CH4 node to both the start and endpoint.
Determine the shortest connection based on the network topology (input: using pg_routing, Kathi Dijkstra algorithm, scikit_learn e.g.).
Compare the length of the shortest route with the actual pipeline length (from the Excel table).
If not within a buffer, find the next nearest point and reapply the algorithm.
If a matching CH4 pipeline is found, reduce the capacity of the pipeline by the capacity of the hydrogen pipeline or deleting whole pipeline
Additional task when matching method for H2- and CH4-pipeline is created:
[ ] Consider the capacity of CH4-pipeline as p_nom_max value for the repurposed H2-pipelines, because capital_costs of repurposed pipelines are much lower than for new-build pipelines
Round about half of the new H2-Pipelines are planed as repurposed CH4-Pipelines
After Implementation of these new H2-Links the repurposed CH4-Pipelines, which the H2-Links will be replace, has to be deleted
[ ] It is necessary to create a method, which is able to match the new repurposed H2-Links with the according CH4-Links for deleting the relevant CH4-Links
First approach to find matches by comparing the distance of start and end points of H2-Links and CH4-Links is not useful, cause the accuracy of H2-Pipelines is not high enough. For large buffer radii, wrong matches between CH4 and H2 pipelines are found, for small buffer radii, a match is only found for a very small proportion of the repurposed H2 pipelines
Ulf's Appraoch (not tried out yet) :
Additional task when matching method for H2- and CH4-pipeline is created:
[ ] Consider the capacity of CH4-pipeline as p_nom_max value for the repurposed H2-pipelines, because capital_costs of repurposed pipelines are much lower than for new-build pipelines