Investigation of vibration data-based human load monitoring system. (May 2021)

• Record Type: Journal Article
• Authors: Sadhu, Ayan; Dunphy, Kyle; Zerpa, Carlos
• Is Part Of: Structural health monitoring. Volume 20:Number 3(2021)
• Page Start: 791
• Page End: 803
• Publication Date: 2021-05
• Subjects: Pedestrian bridges -- human load modeling -- vibration -- walking pattern -- signal processing -- wavelet packet transform -- time-series modeling -- multiple linear regression -- motion sensors; Structural health monitoring -- Periodicals; Structural stability -- Periodicals; Strength of materials -- Periodicals; Nondestructive testing -- Periodicals; Constructions -- Stabilité -- Périodiques; Résistance des matériaux -- Périodiques; Contrôle non destructif -- Périodiques; Electronic journals; 624.17
• DOI: 10.1177/1475921719836254
• Languages: English
• ISSNs: 1475-9217

Abstract

Structural design of flexible footbridges requires a thorough understanding of pedestrian-induced vibration such that their dynamic behavior is accurately predicted. Human-induced vibration creates complicated ground reaction forces that contribute to human–structure interaction in the footbridges. It, therefore, becomes a significant challenge to the bridge designers to accurately estimate the moving pedestrian load on the footbridges during the design phase. This article examines the issues of human–structure interaction in slender pedestrian bridges and aims to analyze the walking pattern of the pedestrian from human-induced vibration data of the bridge. A wavelet-based time–frequency decomposition technique is adopted to extract the walking pattern of the pedestrian followed by time-series analysis of the walking pattern to develop a statistical model of pedestrian-induced vibration. Full-scale testing is conducted on a footbridge to validate the proposed technique under a wide range of pedestrian excitation. An experimental study is conducted to demonstrate the proposed method using the pedestrian's walking on a force plate monitored by video cameras and vibration sensors. Identified walking patterns are then compared with the actual walking patterns measured by motion sensors attached to the test subject.