Civil Engineering March 2021 | Vol 29 No 2

14 March 2021 Civil Engineering Q Q Geometry control by structural defor- mation validation. Structural behaviour monitoring, assessment and verification An extensive monitoring plan was prepared for construction and included system identification by full-scale dynamic testing as well as structural behaviour verification by deformation measurements, requiring continuous modelling and monitoring during all construction stages. Full-scale dynamic testing will be performed when the bridge is in its final position. Deformation measurement was par- ticularly critical during the hanger ten- sioning phase, where the deformation of the arch ribs was measured and compared to theoretical deformation predicted by the calibrated FEM models. The 96 hangers were tensioned in pairs of two so that a symmetrical arch deformation could be achieved. After each tensioning step, the arch-rib and tie-beam displacements were measured and com- pared to theoretical values. The measured displacement was found to closely cor- respond with the predicted theoretical model confirming the designers’ expecta- tion of the structural behaviour. Vertical deflection of the northern and southern arch ribs is presented in Figure 5. Further structural monitoring was done in collaboration with the University of Cape Town’s Concrete Materials and Structural Integrity Research Unit and included strain measuring at vulnerable zones and hanger force measurement. The arch-ribs and tie-beams are each equipped with resistance-type embedded strain gauges (Figure 6a) which are connected to a datalogging system where strains were monitored in real-time during the tensioning operation (Figure 7). These readings verified that adverse concrete behaviour of critical members did not take place during hanger tensioning (van Wijk, 2019). After the tensioning phase, the hanger forces were also checked to ensure that the actual force in the hanger correlated with the forces predicted by the FEM model. Accelerometers shown in Figure 6b were temporarily attached to near midpoint of the hangers and mea- sured their dynamic properties. These properties, together with the individual hanger geometry, were used to determine the tension forces in the hangers. The results confirmed a similar behaviour to what was predicted by the FEM model, but also highlighted the influence of other complexities, such as boundary conditions and material property as- sumptions, that affect the real behaviour of the structure. Figure 4 FEM model Figure 5 Deflection profile after hanger tensioning Deflection (mm) –20 Deflection (mm) Southern arch rib – defection during and after stage Northernarch rib – defection during and after stage Position on deck (m) 0 20 40 60 80 –20 0 20 40 60 80 20 40 60 80 100 120 Position on deck (m) 20 40 60 80 100 120 12 creep 8 creep Profile at 1 February 2021 Profile after Stage 3 tensioning 24 August 2020 Predicted FEM profile after Stage 3 (a) strain gauge (b) accelerometer Figure 6 Instrumentation