A Network Observability Framework for Sensor Placement in Flood Control Networks to Improve Flood Situational Awareness and Risk Management. (May 2022)

• Record Type: Journal Article
• Authors: Farahmand, Hamed; Liu, Xueming; Dong, Shangjia; Mostafavi, Ali; Gao, Jianxi
• Is Part Of: Reliability engineering & system safety. Volume 221(2022)
• Publication Date: 2022-05
• Subjects: Flood control network -- Flood -- Observability -- Infrastructure sensing -- Risk management -- Situational awareness; Reliability (Engineering) -- Periodicals; System safety -- Periodicals; Industrial safety -- Periodicals; Fiabilité -- Périodiques; Sécurité des systèmes -- Périodiques; Sécurité du travail -- Périodiques; 620.00452
• DOI: 10.1016/j.ress.2022.108366
• Languages: English
• ISSNs: 0951-8320

Abstract

Highlights: Observability considerations is neglected in the design of flood sensor networks. A framework for the assessment of flood control network observability is proposed. The framework enables monitoring flood status across critical regions. The framework is tested on flood control network in Harris County, TX. Results reveal insights from observability consideration in sensor network design. Abstract: Monitoring the state of infrastructure systems proactively is crucial to ensure their proper functionality during extreme events. Flood control networks are designed to keep communities safe from inundation. Accurately monitoring the inundation status of flood control components could enhance flood situational awareness and risk management during extreme events. However, the design and placement of sensor networks that collect data to monitor the status of flooding do not often consider the principles of observability. We bridge the gap by creating a framework for the assessment of flood control network observability and determining the minimum number and locations of the flood gauges to achieve maximum monitoring across critical regions. We first delineate critical regions that are needed to be observed, then identify feasible solutions on sensor sets, and finally determine the candidate sensor set based on the importance of the nodes that exist in each sensor set. We tested the framework in the context of Harris County, Texas, as a case study. The results show that the current flood gauge placement is not sufficient for comprehensively monitoring the flooding status of critical areas, nor informed by the network observability principles. Results also offer insights to decision-makers to extend the current flood gauge network or design new flood gauge networks more effectively.