Change of the dynamic characteristics of a pedestrian bridge during a mass event

The human body forms a complex dynamic system with more than one natural frequency in the range of the natural frequencies of pedestrian and stand structures. Recently, for sitting or standing persons a random model of the dynamic behavior of the human body has been obtained. This model allows predicting the range of influences on the dynamic characteristics of the structure. In a simplified approach active persons can be modeled as external loads. While this approach may be sufficient for an activity like jumping, it has been shown already that for a bobbing person some interaction effects may occur. The question arises if also pedestrians are able to influence the dynamic characteristics of the structure they are actually crossing. Observations during a mass event with several thousand persons crossing a 66 m long bridge indicate that the damping capacity of the coupled structure may have increased. The paper presents a time-domain approach to identify the changing dynamic characteristics during the mass event. Basic idea is to use a known background excitation induced by a shaker. The unknown dynamic parameters are identified from a correlation analysis, varying the natural frequency and the damping capacity for a large range of possible systems. The structural responses for these systems are obtained from a step-by-step integration using the observed shaker accelerations as external load. The system leading to the largest correlation leads to the unknown dynamic characteristics. The quality of the approach is verified using a theoretical test case and a laboratory test. Both test cases show promising stability of the estimated dynamic parameters. Finally, the approach is applied to the observations during the mass event. First results are presented demonstrating the potential power of the approach.