VIBRATION-BASED BRIDGE MODAL IDENTIFICATION: FROM FIXED SENSING TO MOBILE SENSING

Author

Nishan Budhathoki, Southern Illinois University CarbondaleFollow

Date of Award

8-1-2025

Degree Name

Master of Science

Department

Civil Engineering

First Advisor

Sen, Debarshi

Abstract

Bridge modal identification plays a key role in structural health monitoring and vibration-based analysis. Traditional approaches rely on fixed-sensor networks and established algorithms such as the Natural Excitation Technique (NExT) combined with the Eigensystem Realization Algorithm (ERA), and Frequency Domain Decomposition (FDD). In this study, these techniques are applied to ambient vibration data collected from the SIUC Campus Bridge over a two-week period for bridge modal identification. Seven dominant modes are identified with stable frequency estimates and consistent mode shapes. However, fixed-sensor networks are often expensive to deploy, limited in spatial coverage and not scalable for continuous monitoring. To address this, a theoretical simulation framework for Crowdsourced Modal Identification using Continuous Wavelets (CMICW) for asynchronous data is evaluated under varying spatial resolutions, traffic speeds, and signal-to-noise ratios. Results show that CMICW maintains high accuracy in modal identification, with MAC values exceeding 97% even at low SNR levels, while traditional methods show degradation in performance under similar conditions. The findings highlight the efficiency of CMICW as a robust, scalable and low-cost approach for modal identifications using mobile sensors such as smartphones mounted on micro-mobility platforms like e-scoters and bicycles. This study supports the transition from fixed-sensor systems to mobile sensing platforms for efficient and scalable bridge health monitoring.