A simplified method to account for human‐human interaction in the prediction of pedestrian‐induced vibrations

Abstract

Summary Vibration serviceability under pedestrian‐induced loading is often the governing criterion for slender footbridge design. During the design stage, the structural response is generally predicted using simplified load models. In these models, neither human‐human interaction (HHI) nor human‐structure interaction (HSI) is taken into account. The present paper starts by characterizing the impact of HHI on the resulting structural response. A widely applied social force model is adopted. The model parameters are adapted to allow for a good approximation of pedestrian flows on footbridges. The HHI‐effect on the resulting crowd‐induced loading and structural response is investigated by means of a numerical study for a relevant range of pedestrian densities. The results demonstrate that an increase in pedestrian density coincides with a decrease in the mean walking speed and the step frequency among the pedestrians in the crowd, as expected. However, no monotonic decrease in the inter‐person variability in step frequency is observed for increasing densities. The inter‐person variability in step frequency is found to also depend on the width of the walkway and the pedestrians' desired room for manoeuvre, which determines the number of parallel pedestrian lanes that can coexist. This paper provides an alternative to the computationally expensive social force model, by translating the HHI‐effect into an equivalent distribution of step frequencies among the pedestrians in the crowd. The results demonstrate that the simplified method allows for a good approximation of the HHI‐effect on the resulting crowd‐induced loading.

Uncertainty quantification and propagation of crowd behaviour effects on pedestrian-induced vibrations of footbridges

“…with two parameters related to the territorial effect 𝐴 𝛼 1 = 9.43 m/s 2 (the interaction strength) and 𝐵 𝛼 1 = 0.35 m (the repulsive interaction range) (Wei et al 2017 andWei 2021). 𝑟 𝛼𝛽 = 𝑟 𝛼 + 𝑟 𝛽 = 0.6 m is the sum of the two pedestrian radii (Helbing et al 1995(Helbing et al , 2000a.…”

“…The radii of the two persons can be assumed as the same, i.e., 𝑟 𝛼 = 𝑟 𝛽 = 0.3 m. 𝑑 𝛼𝛽 is the distance between these two mass centres. 𝜆 𝛼 = 0.82 is a factor to account for the anisotropic nature of the pedestrian interaction (Wei et al 2017 andWei 2021). The anisotropic nature refers to the fact that, the walking person is influenced by more the front pedestrians than the persons behind.…”

Corrigendum

“…In the acknowledgements to the present article, 1 Prof. De Roeck, Prof. Van den Broeck, and Dr. Van Nimmen are mentioned. Given the role that they played in the preparatory research and in the original conception and drafting of the manuscript at the time where the research was done at KU Leuven university, all three of the named persons from KU Leuven deserve to be included as co‐authors.…”

Modal identification and damping performance of a full‐scale GFRP‐SFRSCC hybrid footbridge

“…In the specific case of footbridges, design is often governed by vibration criteria. 2,3 The main aspects included in standards and guidelines in this respect are 4,5 as follows: (i) the need to verify the criticality of the structure's frequencies, (ii) the need to check the possibility of occurrence of lock-in related to lateral vibrations and (iii) the need to verify limit acceleration values under specific pedestrian actions and comfort criteria. Excessive vibration significantly influences the in-service performance of slender structures and their service life.…”