Automatic Completion of Underground Utility Topologies Using Graph Convolutional Networks: Fid-Bau Portal

Automatic Completion of Underground Utility Topologies Using Graph Convolutional Networks

Su, Yang / Wang, Jun / Wu, Peng / Wu, Chengke / Yue, Aobo / Shou, Wenchi

The absence of utility data, particularly about topological information, presents a significant impediment to the efficient management of underground utilities. Previous studies predominantly focus on general attributes such as diameter and material missing, neglecting the imperative issue of insufficient topological information. To address this gap, this study proposes the underground utilities topology completion (UUTC) model based on graph convolutional network (GCN) techniques. A comprehensive evaluation of the proposed model was conducted by performing completion experiments on a real public wastewater network database. This evaluation employed five prominent GCN models while simulating varying missing rates of topological data. The empirical findings indicate that the UUTC model exhibits a substantial advantage over the baseline models, achieving an average completion accuracy of 85.33%. The findings hold the potential to significantly mitigate the expenses associated with manual inspections from incomplete databases.

This research introduces an underground utilities topology completion (UUTC) model, aimed at advancing the management and maintenance of underground utilities such as water and sewage networks. Traditionally, the lack of comprehensive data on these underground utilities’ network topology has posed substantial challenges, leading to inefficient maintenance, unexpected outages, and increased risk of damage during construction activities. The proposed model is capable of accurately predicting the connections within these networks, significantly reducing the reliance on expensive and labor-intensive field surveys. The UUTC model has demonstrated its effectiveness through rigorous testing on real-world data, achieving an impressive average accuracy rate of 85.33% in completing missing topological information. This performance not only surpasses existing methods but also promises considerable cost savings in underground utility management. By integrating data-driven insights and advanced machine learning, the UUTC model offers a practical and efficient tool for improving the safety, reliability, and efficiency of underground utility services, thereby supporting more informed decision-making and strategic planning in urban development projects.