Fatigue or corrosion damage at critical hotspots in offshore and marine structures poses severe threats to the safety of these structures and can lead to substantial financial, ecological or human losses. Accurate prediction of both early stage damage, when the repair cost is low and manageable, and the evolution of the damage, is essential for their sustainable and safe operation. The digital twin, coupled with modelling and sensor data, creates a duplicate of the physical asset in the digital world and allows identification of the critical hotspots and accurate predictions of the damage initiation and evolution under complex loading conditions.

A digital twin for a local detail 

Digital twinning of engineering structures under cyclic actions, e.g. bridges, ships, and offshore platforms, requires a progressive update on the possible crack sizes at critical details. Under cyclic environmental loadings, accurate crack size estimation is essential in evaluating the remaining fatigue life of welded structural components before a repair becomes mandatory. The accurate fatigue assessment depends on the precise fatigue crack initiation and propagation evaluation. This research work proposes an adaptive learning framework to determine and update the crack-front profile at the toe of welded plate joints based on the strain relaxation data. This study determines the crack depth by classifying the nodes in the thickness direction as open nodes on the crack surface and closed nodes on the intact ligament, through a neural network approach (Fig. 1). To update the crack size during the loading history, this research employs the modified bootstrap particle filtering approach (Fig. 2), which entails enhanced adjustment capabilities by imposing additional uncertainty distributions. This approach improves the crack size prediction by absorbing limited measurement data on the strain values or crack sizes (Fig. 3).

For more details, please contact:
Assoc Prof Qian Xudong
Email: qianxudong@nus.edu.sg