Quasi-Self-Powered Piezo-Floating-Gate Sensing Technology for Continuous Monitoring of Large-Scale Bridges

Abstract

Developing a practical framework for long-term structural health monitoring (SHM) of large structures, like a suspension bridge, poses a major challenge. The issue has been exacerbated by the fact that many of the world’s bridges are reaching the end of their initially designed service lives. The next generation of bridge SHM technology not only needs to continuously monitor and issue early warnings prior to costly repair or catastrophic failures, but also needs to gauge the effects of rare events like earthquakes or hurricanes on structural conditions. Current battery-powered SHM methods use periodic sampling with relatively long sleep-cycles to increase a sensor’s operational life. However, long sleep-cycles make the technology vulnerable to missing out on or incorrectly measuring the effect of rare events. To address these practical issues, we present a novel quasi-self-powered Piezo-Floating-Gate (PFG) sensing solution for long-term and cost-effective monitoring of large-scale bridges. The approach combines our previously reported and validated self-powered PFG sensing technique with ultra-low-power long-range wireless communications. The physics of PFG sensing continuously captures and stores cumulative, local information of the bridge dynamic loading condition on a floating-gate based non-volatile memory. Using extensive numerical and laboratory studies we demonstrate the capabilities of the PFG sensor towards predicting structural condition. We then present a system level design that adapts PFG sensing for SHM in bridges. A challenging aspect of SHM in bridges is the need for long-range wireless interrogation, as many portions of the structure are not easily accessible for frequent inspection and portions of the bridge cannot be frequently taken out-of-service. We show that by combining PFG sensors with a small battery and optimizing the quasi-self-powered system for sporadic long-range active wireless transmissions, the system can easily achieve continuous operating lifespan exceeding 20 years. The efficiency of the proposed method is verified in a case study of the Mackinac Bridge in Michigan, the longest suspension bridge across anchorages in the Western Hemisphere. Associated data interpretation systems integrating deterministic, machine learning and statistical methods are presented; furthermore, limitations, challenges and future directions for widespread field deployment of the proposed SHM framework are discussed.