Distribution widths for concrete slabs

Abstract

ABSTRACT The distribution width of sectional forces and moments for a one-way concrete bridge deck highly influences the amount of reinforcement required for the deck. In engineering practice, two calculation methods are used: simplified analytical calculations and finite element (FE) analysis. Compared to FE analysis, the analytical calculation is more convenient to use but due to the lack of accurate guidelines, it is too conservative and hence overestimates the reinforcement. This thesis focuses on providing a more accurate expression of distribution width for the analytical calculation by utilizing FE parametric study to enhance the accuracy of the analytical calculation. The approach to determine the distribution width in this thesis is by comparing the sectional force from a linear 2D shell FE model to an analytical calculation model. The ratio between sectional forces of these two models was the approach to finding the new expression. Based on this approach a full parametric study was required where multiple models were analyzed, and results were collected. Following that, the expressions of the distribution widths were established based on all the results collected from these various models. To simplify the task, a specific model was chosen representing a common bridge deck, and this model was referred to as the “base model”. The whole approach was tested on the base model first, and several problems and limitations were identified. One of them was the constrain caused by the cantilever in the FE model for a load at the mid-span, which led to a mismatch of the zero-moment position between the two models. Consequently, a “dummy load” was introduced into the analytical calculation model. Another problem was the extensive data processing to include the torsional moment in the design;this was avoided by choosing the average width as zero. Graphical curve fitting was utilized to create the expressions of the distribution widths and other related calibration variables such as dummy loads. Furthermore, results comparison envelopes between the new expression and the raw FE results were used to check the accuracy of the expressions. Finally, recommendations for use in engineering practice is given, and several issues are discussed such as the “cutoff” region for shear force expression, and the influence of the support stiffness and slab orthotropy on the new expression.