OPTIMUM DESIGN OF FRP LAMINATED BRIDGE DECKS HAVING DIFFERENT CORE CONFIGURATIONS

Abstract

FRP laminated composite structures are gaining importance due to their high relative stiffness, high strength, light weight and enhanced durability properties. FRP laminated composites represent efficient and favorable solutions to the deficiencies of many traditional materials and have a great potential to integrate into the bridge infrastructure. The increase in deterioration of bridge decks and culverts is a large-scale worldwide problem. Maintaining the existing bridge infrastructure network and adapting it to new capacity requirements has become one of the most challenging tasks for today’s engineers. Bridges designed and built only a few years ago are now subjected to traffic loads well above the design values. At the same time, it has become evident that the durability of bridges is not always guaranteed, even for relatively recent constructions. It is, therefore, necessary to intervene with feasible and efficient methods to upgrade, repair or strengthen the existing bridges while preserving, its traffic bearing capacity. The main problem in the use of FRP laminated bridge deck structures lies with the higher initial cost though the cost should drastically come down with rapid use of this material. However, proper analysis, design and optimization of the FRP laminated bridge deck structure would still make it economically feasible. The structures made of FRP materials also have some design difficulties associated with their complex material properties, presence of multiple design variables and objectives which lead to a complex design problem. Although many experimental and numerical studies have been conducted, yet there is no well-defined optimum design methodology available in the literature for FRP laminated composite structures including bridges. The codes and standards also do not have explicit provisions for designing bridges using composite materials for full and widespread utilization. Therefore, design and optimization of FRP laminated composite structures including bridges is an interesting challenge in the field. The process of design optimization involves many stages. In the first stage it is required to select some initial geometric dimensions to start with the basic design of FRP laminated bridge deck like structures. This can be achieved by using some simplified (e.g., 2D) modelling techniques by satisfying the design constraints to save a substantial portion of computational time required otherwise in a rigorous analysis (e.g., 3D analysis). In the next stage, a more accurate rigorous analysis should be done in limited range of parametric variation of the design variables (e.g. thickness of different bridge components) by suitable optimization schemes. ii Finally, the optimized design configuration should be checked for all possible deflection and stress criterion. It is also essential to experimentally verify some of these theoretical results to arrive at the final recommendations. In the present study, two different bridge deck configurations, namely, sandwich core and web core are considered which are subjected to IRC 70R tracked vehicle load with impact as the live load in addition to self weight. The structural optimization of the geometric dimensions of these two bridge deck configurations are performed separately in different stages. It is observed that the designs of FRP laminated composite bridge deck panels are generally governed by deflection as the elastic modulus of the material is relatively low. As there is no code available in India for the design of FRP composite bridge panels, the deflection criteria (i.e. span/800) specified in American Association of State Highway and Transportation Officials (AASHTO) guidelines are used in the present study. Also deflection is considered as the basic design constraint in all design optimization schemes for sandwich core as well as web core bridge panels. The objective function is considered as the volume of the bridge deck components which actually minimizes the cost. The optimum solution (i.e. least volume) is obtained by satisfying all design constraints such as buckling and stress failure along with the basic deflection criteria. A 2D FE model based on refined higher order shear deformation theory (RHSDT, i.e., Zigzag theory) is proposed for simple analysis of FRP sandwich core panels to get the preliminary geometric dimensions at the initial stage based on a novel homogenization technique. Similarly, a 2D FE model has been developed for FRP web core bridge deck structure based on First Order Shear deformation theory (FSDT) for initial design. The preliminary designs obtained from these FE models are then used as initial data for optimization. Based on these preliminary configurations of FRP bridge decks, numerical modeling is carried out using Graphical User Interface (GUI) of ABAQUS (2008). The model prepared in ABAQUS (2008) is further used to generate python script. Typical multi-looping are incorporated in the python script to make a parametric model for optimization and the modified python scripts are used to generate various results of different trial solution required for the optimization. The total volume of the FRP used for the bridge deck is considered to be the objective function to be optimized. Deflection as well as buckling and failure stresses are iii considered as the design constraints for optimization of FRP composite laminated structures. The basic optimization methodology carried out as mentioned above is named as Gradient Search Optimization (GSO). This basic optimization method is subsequently enriched by a modified optimization technique based on Response Surface Method (RSM). The RSM method has shown a greater efficiency over GSO in terms of computational time and efforts. The optimization of FRP bridge deck structures based on RSM is performed using two algorithms (viz. fmincon and Genetic Algorithms) available in MATLAB (2011) to get optimum design configurations. The structural configurations obtained by the above mentioned theoretical procedure are verified by experimental studies before generating large number of new results. The optimized results thus obtained are also presented in the form of design charts.