Abstract

The main indicator of damage in concrete structures are existing cracks. In this regard, modern wireless sensing networks can be used to provide continuous measurements of crack displacements, strains, corrosion indicators, temperature and other quantities. In addition, drones equipped with high-resolution cameras and coupled with artificial intelligence (AI) algorithms can provide accurate maps of the cracks and damage in the structure. However, despite the advances in modern technology, a key question remains: how to transform the monitoring data from existing cracks into useful knowledge about the safety of the structure?

To address this question, a kinematic modelling approach has been developed at the University of Liege for various concrete members and regions. This approach is based on an explicit kinematic modelling of the critical cracks (crack widths and slip), and also includes equilibrium conditions and constitutive relationships for the load-bearing mechanisms in the member. The seminar will discuss how this approach can use crack data as an input to predict the complete behaviour and residual load-bearing capacity of cracked members. Furthermore, for members with inadequate residual capacity, a retrofit solution using ultra-high performance fiber reinforced concretes (UHPFRC) will be discussed. It will be shown that the crack-based modelling approach is also well suited for evaluation of the effect of UHPFRC strengthening.