Developing data science for structural safety analysis and pre-warning in civil engineering

Abstract

The development of data science in civil engineering has benefited from the rapid advances in sensor technology as well as data acquisition and storage. In contrast to traditional analysis and evaluation based on periodic inspection and full-scale test, the structural safety analysis and pre-warning can be achieved directly through the analysis of the design data of the newly built bridge and the monitoring data & test data of the bridge in service. Specifically, structural geometric data (length and cross-sectional area etc.), physical response data like displacement and stress, and vibration response data, such as acceleration and frequency, as well as the influence of the environment, e.g., temperature and humidity, must all be taken into account. Furthermore, the different sensitivity of different response data, which in turn affects structural safety analysis and pre-warning accuracy, is one of the current frontier sciences, i.e., the problem of multi-source (different response) data. It is expected that the development of data science will have very important theoretical research value and engineering practice significance for safety analysis and pre-warning in civil engineering, and is expected to bring new prospects for academia and industry.