PSI - Issue 64

Nicoletta Bianchini et al. / Procedia Structural Integrity 64 (2024) 352–359 Bianchini N., Sabra Z, Green K.. Wrigth, R. Structural Integrity Procedia 00 (2024) 000 – 000

357

6

Table 1. List of the numerical models and their characteristics

Identifier Description

Loading conditions as per CS 454 and CS 458

M1

Reference model Including defects Flooded condition Two rings

Model with reference properties (Table 2) and dry condition

M2

Model with reference properties (Table 2) and dry condition Loss of mortar within localized joints (see Fig. 6 red lines) Model with flooded properties (Table 2) up to the level of the deck (see Fig. 6 water level) Model with reference properties (Table 2) and dry condition. Two separate rings and accounting for ring separation.

Normal traffic, HB and Special vehicles

M3

M4

In all the scenarios, cohesion is set to null , and the contribution of model dispersions and horizontal pressures are taken into account. There was no evidence of structural backing in the available existing information and therefore was not considered in the assessment. As detailed in Table 2, M1, M2, and M4 share the properties which are recurrent for masonry arches of South of England. In general, the arch profiles were modelled based on the geometrical data obtained during the topographical survey to produce a realistic arch shape for the 2D analysis. While M1 can be considered as a baseline reference model, M2 includes the loss of mortar in the joints and M4 includes two arch rings. Regarding M2, given the detailed available geometric information, accounting for defects like loss of mortar and potential plastic hinge on the western span was possible. In general, a loss of 40-100 mm of mortar was considered in specific locations, likely due to the erosive effect of the water. On the other hand, M4 considers the role of a double ring, where the second one has a thickness of 255 mm constant along the two barrels. A standard value for the brick masonry was assumed for M1, M2, and M4, however for M3, the unit weight of masonry and backfill material were varied. The scenario at Waverley Mill bridge included the condition of variable masonry and backfill unit weights represented by M3 model. This is due to the permanent/constant flooded state of the structure to the level of the the crown (see Fig. 5, Fig. 6). Thus, M1 and M3 properties differ because of the unit weight of the materials in M3 considers the Archimedes force contribution. In particular, the unit weight of submerged materials was reduced using the buoyancy ratio calculated based on the dry unit weight and the unit weight of water. The buoyancy ratio typically ranges between 1.6-1.8 for historic masonry bridges.

Fig. 6. Scheme of the numerical models M1, M2 M3 and M4 in RING environment.