Scour around bridge piers remains the leading cause of bridge failure induced in flood. Floods and torrential rains erode riverbeds and damage cross-river structures, causing bridge collapse and a severe threat to property and life. Reductions in bridge-safety capacity need to be monitored during flood periods to protect the traveling public. In the present study, a scour monitoring system designed with vibration-based arrayed sensors consisting of a combination of Internet of Things (IoT) and artificial intelligence (AI) is developed and implemented to obtain real-time scour depth measurements. These vibration-based micro-electro-mechanical systems (MEMS) sensors are packaged in a waterproof stainless steel ball within a rebar cage to resist a harsh environment in floods. The floodwater-level changes around the bridge pier are performed using real-time CCTV images by the Mask R-CNN deep learning model. The scour-depth evolution is simulated using the hydrodynamic model with the selected local scour formulas and the sediment transport equation. The laboratory and field measurement results demonstrated the success of the early warning system for monitoring the real-time bridge scour-depth evolution.
This study proposes an information-theoretic structural health monitoring (SHM) system based on multi-scale entropy (MSE) and multi-scale cross-sample entropy (MSCE). By measuring the ambient vibration signal from a structure, the damage condition can be rapidly evaluated via MSE analysis. The damage location can then be detected by analyzing the signals of different floors under the same damage condition via MSCE analysis. Moreover, a damage index is proposed to efficiently quantify the SHM process. Unlike some existing SHM methods, no experimental database or numerical model is required. Instead, a reference measurement of the current stage can initiate and launch the SHM system. A numerical simulation of a four-story steel structure is used to verify that the damage location and condition can be detected by the proposed SHM algorithm, and the location can be efficiently quantified by the damage index. A seven-story scaled-down benchmark structure is then employed for experimental verification. Based on the results, the damage condition can be correctly assessed, and average accuracy rates of 63.4 and 86.6% for the damage location can be achieved using the MSCE and damage index methods, respectively. As only the ambient vibration signal is required with a set of initial reference measurements, the proposed SHM system can be implemented practically with low cost.