Kim, J; Kamari, M; Lee, S and Ham, Y (2021) Large-scale visual data-driven probabilistic risk assessment of utility poles regarding the vulnerability of power distribution infrastructure systems. Journal of Construction Engineering and Management, 147(10): 04021121, ISSN 0733-9364
Abstract
Inspecting and assessing existing utility poles has become increasingly important for reducing the vulnerability of power distribution infrastructure systems in disaster situations, which can enhance community resilience. Although vision-based systems have been applied to detect faults in power distribution infrastructures, little research currently exists on assessing component- and network-level failures of utility poles based on their geometric and environmental information. This paper aims to propose a new data-driven approach to support risk-informed decision-making for utility maintenance under extreme wind conditions. Large-scale open-source imagery from Google Street View is used to assess geometric properties of utility poles (i.e., leaning angle). Then the failure probability of utility poles is analyzed under varying conditions (e.g., age, leaning angle, and wind loads) in a three-dimensional virtual city model. The proposed method is tested through case studies in Texas to (1) validate an algorithm for estimating leaning angles of utility poles and (2) understand the progress of failures of leaning utility poles from a network perspective. The outcomes of the case studies demonstrate that the proposed method has the potential to leverage large-scale open-source visual data to assess the vulnerability of utility pole networks that may lead to cascading failures in power distribution infrastructure systems. Based on the proposed virtual environment, the method is expected to enable practitioners to facilitate risk-informed decision-making against disaster situations, which creates an opportunity for prioritizing maintenance tasks regarding power distribution infrastructures.
| Item Type: | Article |
|---|---|
| Uncontrolled Keywords: | computer vision; deep learning; google street view; machine vision; power distribution infrastructure systems; probabilistic risk assessment; vulnerability |
| Index terms: | community resilience, wind load, case study, deep learning, risk assessment, prioritizing, vulnerability, risk-informed, computer vision, machine vision, estimating, street view, practitioner, virtual environment, decision-making |
| Subjects: | financial risk, environmental hazards, financial and cost management, social equity, computer vision, artificial intelligence, urban design, decision analysis, practitioner, virtual reality, data collection methods, structural engineering |
| Topics: | Engineering Principles, Sustainability, Research Practice, Roles and Professions, Risk Management, Digital Applications, Cost Management, Urban Studies |
| Descriptive scope: | 3 PCE |
N.B. Descriptive scope is a count of how many of the five facets of empirical research are indicated by the words used in title, abstract and keywords. It is not intended as a judgement on the research; merely a count of the kind of word we would expect to indicate Phenomenon, Concepts, Theoretical framing, Empirical techniques, Analytical techniques. If all five are present, then a code of “5 PCTEA” will indicate this. If you feel the coding for this record is questionable, we welcome discussion around the terms we matched or the way we categorized them. The facet you would expect may not be coded, or a facet may be coded inappropriately. This can also bear on a larger question, of which facets should be treated as defining in construction management research. Please get in touch, and we will look at it. More details here