Enhancing sensor data reliability in structural health monitoring systems using digital twin technology

Abstract

Structural health monitoring (SHM) systems are intended to collect continuous, current information on the condition of civic buildings, aircraft, and industrial gear. These structures may decay over time as a result of slow processes such as corrosion and fatigue, as well as unexpected catastrophes such as natural or man-made disasters. It is important to note, however, that the components of SHM systems may degrade after installation, resulting in incorrect information being supplied to decision-makers. To address this issue, digital twin (DT) technology has been utilized to provide an accurate representation of the building's actual state. Additionally, a weighted-Bayesian belief network (W-BBN) has been integrated into the framework to assess the building's condition. The framework also includes an example of maintenance and inspection throughout the lifespan of the structure, incorporating a sophisticated model of the SHM system to evaluate the level of risk. To show the proposed framework's operations, a numerical analysis and technique demonstration are offered, utilizing the building's smart sensors as a reference. The findings reveal that the unique SHM technology inspired by DT increases the identification of building health in an efficient and automated way.