Corrosion Identification, Location and Assessment of Prestressing Tendon Anchorages in Suspension/Cable-Stayed Bridges: Integrating Image Recognition and Machine Learning with Acoustic Emission Method

Abstract

The intensive development of transport infrastructure in Poland and worldwide has led to a rapid increase in the number of engineering bridge structures. Prestressed concrete is currently the most commonly used structural material (43.4%). One advantage of prestressed concrete is the ability to achieve considerable span lengths; consequently, the percentage share of the surface area of such bridge structures is even greater, reaching 58.2% by the end of 2017. Although visual checks are feasible for tendon components, assessing prestressing force and diagnosing cable issues (like corrosion, grout condition, and voids) in prestressed structures is a significant technical and research challenge. This paper introduces a monitoring approach, using periodic inspections or continuous electronic monitoring, to evaluate their condition. This approach employs a novel measurement system that identifies anomalies by recognizing patterns in acoustic emission signals. Furthermore, the pattern classes have also been correlated with crack opening widths, allowing for tracking the influence of crack development on the structural stiffness and monitoring other processes such as corrosion and slippage at the anchorage.