Civil engineers are presently faced with the challenge of strengthening and repairing many existing structures to assure or increase their structural safety. The reasons for this include changes in the use of structures, and increased traffic loads on bridges. In Iraq, for example, several highway bridges needed to accommodate increased axle load during the transportation of huge turbines for electricity generating stations. The requirement for structural strengthening and repair methods is, however, driven by the worldwide need to ensure the safety and sustainability of our aging infrastructure which is deteriorating at a rate faster than it can be renovated. The ever increasing damage caused by environmental effects and the corrosion of steel and deterioration of concrete, reduce structural safety and lead to disruption for the users, which can have serious economic consequences. In a plate girder bridge, the plate girders are typically I-beams made up from separate structural steel plates (rather than rolled as a single cross-section), which are welded or, in older bridges, bolted or riveted together to form the vertical web and horizontal flanges of the beam. The two primary functions of the web plate in a plate girder are to maintain a relative distance between the top and bottom flanges and to resist the induced shear stresses. In most practical ranges of plate girder bridges’ spans, the induced shear stresses are relatively low compared to the bending stresses in the flanges induced by flexure. As a result the web plate is generally chosen to be much thinner than the flanges. The web panel consequently buckles at a relatively low shear force. For steel girder structures dominated by cyclic loading, as is the case with repeated vehicle axle loads on bridges, this can lead to the so-called ‘breathing’ phenomenon; an out-of-plane buckling displacement that can induce high secondary bending stresses at the welded plate boundaries. In the current work, a novel FRP strengthening technique using bonded shapes is applied to resist these out of plane deformations, and hence reduce the breathing stresses, and improve the fatigue life of the plate girder which is very different to the majority of applications of FRP strengthening that exploit the FRP for its direct tensile strength and stiffness. The objective of the current experimental programme is to strengthen thinwalled steel girders against web shear buckling using a corrugated CFRP or GFRP panel bonded externally along the compression diagonal of the web plate. The programme was divided into three main phases, including: (1) the development of a new preformed corrugated FRP panel, and (2, 3) testing its performance in two main experimental series. The initial series involved tests on 13 steel plates strengthened with the proposed preformed corrugated FRP panel and subjected to in-plane shear loading using a specially manufactured “picture-frame” arrangement designed to induce the appropriate boundary conditions and stresses in the web plates. This initial test series investigated the performance of different forms of strengthening under static load, in preparation for another series of cyclic tests to investigate their fatigue performance. The test variables included FRP type (CFRP or GFRP), form of FRP (closed or open section), number of FRP layers, and orientation of GFRP fibres used to produce the FRP panel. In the second series, six specimens were manufactured to simulate the end panel of a plate girder. These were strengthened with the optimized FRP panel from the initial series and tested for shear buckling under repeated cyclic loading with a stress range 40-80% of the static ultimate capacity. A considerable increase in the stiffness of the strengthened specimens is evident in the observed reductions of the maximum out-of-plane displacement. The stiffness of the strengthened specimens is assessed to be increased by a factor ranging between 3 to 9 times the stiffness of the corresponding unstrengthened specimen, depending upon the type of the FRP panel used and the aspect ratio of the tested specimens. The breathing phenomena is also significantly reduced, consequently the surface, membrane and secondary bending stresses are reduced. The 45° strengthening scheme succeeded the best both in reducing the breathing stresses and increasing the ultimate shear capacity of the specimen by 88%. Fatigue analyses indicated that the proposed strengthening technique is able to considerably elongate the life expectancy of the strengthened plate girders by a factor ranging between 2.5 and 7 depending on the applied cyclic load amplitude. In addition, the proposed strengthening technique did not show any debonding or delamination under both static and cyclic loading which makes it a good candidate for strengthening thin-walled structural members, especially, when ductility is a concern. In fact, the proposed strengthening technique succeeded in improving the energy absorption capacity of the strengthened specimens by a factor ranging between 1.5 and 2.5 times the corresponding control specimen which means that the ductile failure type associated with shear buckling of steel plate girders is not only maintained, but it was improved as well. This type of ductile failure is not common in other types of FRP strengthening techniques. Finally, a geometrical and material non-linear finite element model is presented both for the steel and composite sections which showed very good correlation with test results and was capable of predicting both the strength and deformational behaviour of the tested specimens. This numerical model is used for a parametric study to support the proposed design method.
| Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:726529 |
| Date | January 2016 |
| Creators | Al-Azzawi, Zaid Mohammed Kani |
| Contributors | Stratford, Timothy ; Rotter, Michael |
| Publisher | University of Edinburgh |
| Source Sets | Ethos UK |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
| Source | http://hdl.handle.net/1842/25453 |
Page generated in 0.0034 seconds