Large-scale generation of geo-referenced power distribution grids from open data with load clustering

Abstract

The availability of real power distribution grid data is often restricted due to privacy concerns and the lack of digitized representations, limiting spatially-resolved assessments of these systems. This inaccessibility has motivated the development of methods for generating synthetic grids. However, existing methods face challenges such as computational intractability for large-scale zones, restrictive topological assumptions, insufficient representation of electrical components, and inadequate consideration of geographical constraints. This work addresses the challenges by developing a model for the large-scale generation of synthetic geo-referenced low- and medium-voltage grids using publicly accessible data. It comprises a geographic load clustering algorithm, a procedure for generating graphical grid layouts, and a method for selecting operational topologies and line types. The model’s effectiveness and computational performance are demonstrated by generating synthetic low- and medium-voltage grids for Switzerland, with all generated grids made openly available.