Optimisation of a high strength concrete plate bridge

Since the first high strength concrete mixtures were designed in the 1970s, a lot of research is performed. Despite the recommendations that were written, full legislative coverage of high strength concrete remained unavailable. The new Eurocodes are also valid for high strength concrete, so the question rises how high strength concrete can be used in the civil engineering sector. This Master’s thesis is therefore aimed at the question which (combinations of ) cross sections can be used to create an optimal design for cast in situ high strength concrete bridges. From a consideration of the dominant loads on a concrete plate bridge, it followed that for bending moments the governing cross section is required over a small length of the entire bridge. This also applies to shear forces. It was also found that the self weight of the bridge forms a large part of the total load on the bridge. Research into the optimal cross section for the three main actions (normal force, bending moment and shear force) learned that for normal and shear forces solid cross sections are optimal. For bending moments on the other hand, I profiles perform best since the material is concentrated at the top and bottom of the cross section. Attempts to improve the performance by including a high strength concrete in a normal strength concrete cross section yielded little success, mainly due to the small difference in stiffness between both materials. Both investigations lead to the development of two strategies for optimisation of the solid, flat plate bridge. The first strategy is to vary the cross sectional height along the length of the bridge. This reduces the bending moment capacity at the locations where the maximum capacity is not required. The other method is to apply the optimal cross sections at the positions where the demand for the corresponding capacity is large. Using the I section at positions where bending moment capacity is required yields a plate with rectangular voids in longitudinal direction. After the benchmark design (the solid flat plate) is optimised, the designs for each strategy were optimised as well. During the optimisation process, as much material as possible is removed while maintaining the designed shapes, until the first structural check failed. The optimal variant that scored best was a one metre high plate with 20 voids. Compared with the initial cross sectional area of the bridge, 33% is replaced by voids. When comparing the cross sectional area with the benchmark design, the reduction is almost 40%. Due to the reduction of the plate thickness by 100 millimetres, the span to plate thickness ratio is increased from 35 (rule of thumb) to 42. During the tender process the durability and sustainability of designs becomes more important. Hence the MKI method as used in Rijkswaterstaat’s DuboCalc software is used in order to assess the environmental impact of the various designs. This method incorporates a number of environmental factors and translates this to environmental costs for each construction material (MKI value). Compared with the benchmark design, the optimised design reduces the MKI value by 37%.Structural and Building EngineeringDesign and ConstructionCivil Engineering and Geoscience