Generating new electricity cable trajectories – a methodological study

Abstract
For the advancement of the energy transition, the installation of new electricity infrastructure is of great importance. In particular in the Netherlands, the network is lagging behind; it cannot keep up with the demand for new connections. This can be described as transport scarcity, or transportschaarste in Dutch. Earlier research on the allocation of new electricity cable trajectories has primarily focused on using a least cost path analysis using a cost raster. This thesis takes a different approach; it uses the road network to guide the least cost path analysis on the cost raster within a new workflow for the generation of electricity cable trajectories.

This workflow entails three algorithms to find multiple possible paths on the road network that are constrained to the maximum length of cables. These algorithms result in sets of paths on the road network. These paths reduce the cost raster and guide a least cost path analysis on the reduced cost raster to the areas surrounding roads, which are most suitable for new electricity cables. In this way, multiple options for electricity cable trajectories are created.

This analysis worked: solutions were found. It is computationally intensive, where for two of the algorithms the sets of generated paths were large, consisting of hundreds of paths. The third algorithm was less computationally intensive and also generated less paths. However, this algorithm does not compute all possible paths and might miss promising paths and consequently advantageous trajectories which the first two algorithms do compute.

The results of this analysis show that this workflow offers potential for the future generation of the trajectories of new electricity cables, but there are some points of interest. An example is the choice of algorithms, which is a trade-off between the computational intensity and the comprehensiveness of generated trajectories. Additionally, future research should amongst others focus on the used cost raster and the usage of different, possibly more efficient algorithms.