Advanced risk assessment framework for land subsidence impacts on transmission towers in salt lake region

Abstract

To ensure the continuous transmission of clean energy for electricity, transmission towers often traverse geologically challenging terrains. This is particularly evident in the Salt Lake region, where the convergence of unique geological attributes and environmental stressors leads to recurrent land subsidence incidents, jeopardizing the integrity of power lines. In light of the prohibitive costs associated with indiscriminate remedial actions, precision in risk assessment and management is imperative to streamline these interventions. The objective of this study is to propose an accurate and effective framework for assessing the risk of land subsidence for power transmission towers. To achieve this, by utilizing the Small Baseline Subset Interferometric Synthetic Aperture Radar (SBAS-InSAR) technology and cross-validation from field investigations, representative sample disaster datasets of surface deformation points of transmission towers are obtained in the first. Two optimization approaches for susceptibility modeling were employed to further enhance the susceptibility model for land subsidence and obtain high-precision susceptibility results along the transmission towers. Then, combined on-site damage investigation with SBAS-InSAR technology to conduct subsidence vulnerability analysis of transmission towers. Finally, the ultimate outcome of this process was the development of a risk assessment matrix based on spatial probability (susceptibility) and damage scales (vulnerability), aimed at more accurately assessing land subsidence risk. This framework could identify the spatial distribution of subsidence susceptibility in transmission towers exposed to potential damage, which could be followed by a more detailed transmission tower-based vulnerability and risk assessment. The method has been successfully applied in the Salt Lake area, and the susceptibility results based on the Grey Wolf Optimization-Random Forest (GW0-RF) model have shown the best modeling effect, and the area under of receiver operating characteristic curve (AUC) = 0.95. The vulnerability results show that 105 towers have been most severely damaged by land subsidence The risk distribution map shows that the central part of the Salt Lake area is the core zone for future land subsidence risk management. A total of 101 towers are in the high and very high risk level. The insights garnered from this investigation furnish a comprehensive understanding of the risk landscape for transmission towers, underpinning future-focused design, planning, and management strategies.