Utility geographic information systems store valuable spatial data on distribution networks, but raw GIS databases frequently contain structural defects that prevent their direct application to electrical analysis. Geometric discontinuities, missing or implicit asset relationships, and incomplete node-to-node connectivity are common obstacles. This study presents a practical screening methodology that converts GIS geospatial layers into electrical graph representations and systematically audits their fitness for use in operational and planning studies.
The framework operates in stages. First, it translates GIS feature layers into a graph structure suitable for topological analysis. It then audits key distance metrics—measuring how far loads sit from their assigned poles and how far line segment endpoints deviate from known network nodes. Sensitivity analysis examines how the network structure changes as connection tolerance thresholds vary, simulating the effect of different assignment rules. The methodology assigns each load a confidence score reflecting data certainty, and evaluates transformer loading under a realistic demand scenario. Finally, synthetic perturbations of line geometry test the robustness of the derived topology.
Applied to a 38-block urban district, the case study analyzed 429 loads across 155 poles and 13 transformers connected by 800 LV segments. Results showed that most loads (96.97%) remain stably assigned to poles across reasonable tolerance ranges. However, the number of virtual nodes created during graph construction was sensitive to connection tolerance, varying from 817 nodes at strict tolerances to 439 at relaxed thresholds. Confidence classification revealed that 32% of loads earned high-confidence assignments, while 5% required review. Four transformers exceeded 80% loading under residential demand assumptions, flagging potential capacity issues.
The contribution is significant: utilities gain a transparent, reproducible procedure for quality assessment before deploying GIS data in hosting-capacity studies, distribution planning, or advanced controls. The methodology does not reconstruct the true physical topology but rather quantifies data reliability and guides remediation priorities.
