Real-Time Ground Vibration Estimation on Construction Site

Abstract

Construction-induced ground vibrations present a significant concern, as vibration can cause health risks to workers and structural damage to surrounding buildings. Given the dynamic nature of construction activities, real-time monitoring and management of vibrations become critical. However, current ground vibration management methods rely on predictions based on pre-construction information or measures at a certain time frame at a certain location. These methods often fail to incorporate real-time data, hindering their ability to respond to dynamic changes and reflect real-world conditions. To overcome these challenges, the study introduces a novel method for estimating ground vibrations using a customized spatial interpolation model and distributed vibration sensors on the site. The proposed method begins with the collection of vibration data from sensors installed across a construction site, followed by the creation of a comprehensive dataset incorporating sensor locations and sensing time. The customized spatial interpolation model, specially designed for construction sites considering spreading patterns of ground vibration, estimates vibrations at the overall construction site and visualizes the results on a point cloud map. The effectiveness of the proposed method was validated at a road construction site involving a pile-driving activity, one of the most vibration-causing activities. With six sensors for interpolation to predict each spot’s vibration level and four for validation of the predicted values with real sensed data, the experiment yielded 90.5% accuracy in estimating the overall vibration of the site, underscoring the method's monitoring precision and reliability in real time. Consequently, the proposed method is expected to facilitate immediate modifications and corrective measures for vibration-inducing activities at construction sites. The method will be further enhanced with higher accuracy by incorporating sophisticated finite element modeling techniques