As many bridges in the Netherlands are past their design-life and yet still have well-functioning steel superstructures, their bridge decks are often due to be replaced. FRP panels are promising alternatives to concrete decks for these renovation works due to their low density, fast on-site construction time, and excellent acoustic and thermal properties. Many of these bridges are originally designed with hybrid interactions occurring at the connection level. Multiple connection types exist to ensure hybrid interaction with FRP decks. However, there is a gap in this field concerning a rapidly demountable connection able to transfer shear forces at girders with oversized holes. The injected reinforced resin connector is designed and aimed at achieving hybrid interaction through combined injection and preloading of the connector. The aim of this thesis is to find a viable design for this connection, and to test the connection under local compressive wheel loads. be able to design the connector and test it according to standards as expected to occur in real bridge applications, a FE-model of the bridge Nieuw-Vossemeer is made to assess expected loadlevels in the connection. The result is a combined list of loadlevels and geometrical tolerances that the connection will have to be able to sustain or perform under. Based on the resulting design requirements a main design is proposed. Here a steel bolt is embedded in the bridge deck and held in place by injection with reinforced resin. A steel plate separates the bridge deck from the girder flange, and creates a full steel clamping package. Multiple design alternatives are tested by the use of FE-models, where the variable is the type of steel plate used between the bridge
deck and the girder flange. The most viable design to continue experimental tests with is selected by its stiffness, ultimate resistance, and a qualitative assessment regarding expected cyclic endurance. The best performing concept is a steel plate with a spherical bottom, a diameter of 150mm and maximum thickness of 20mm. Extra FE-analyses are performed on a local scale, and take into account positional deviations from the most ideal situation. In addition a series of two static and two cyclic experiments has been performed. Apart from finding their respective resistances, the failure modes are assessed and compared together with the numerical analysis. The static tests show an ultimate resistance of 501kN, the numerical analysis shows a resistance of 498kN. Both the static tests and the numerical analysis failed ultimately due to delamination in the bottom facing of the FRP, the cyclic tests failed under a similar failure mode in the bottom FRP flange. On loading ranges with R = 0.1 it is found that the primary failure mode is a horizontal crack opening up in the bottom FRP flange, which propagates into the webs. The connection withstood 2 million cycles at 9kN - 90kN without damage. Subsequently, two different loadranges were applied to the two different tests, which failed after 394 thousand cycles at 13.5kN - 135kN and 34 thousand cycles at 18kN - 180kN. It is concluded that the spherical plate is able to function as a coverplate and withstands rotations, displacements and the required loadlevels as expected in bridge applications. The primary recommendation is to continue performing tests with a similar plate with a thickness of 15mm, as the current design overshoots the required resistance significantly.
The results show that the design is well able to sustain the required local compressive wheel loads, both in ultimate and cyclic resistance. With the ability to (de)mount these FRP panels quickly on-site, two main goals have been reached. Firstly, this brings the use of FRP panels in infrastructure closer, leading to reduced loads onto steel superstructures of existing bridges (and hence an extended lifetime). Secondly, a fundamental step has been set towards modular bridge design, where during failure only specific panels of the system have to be replaced. Together these developments can lead to a reduced impact of the construction industry on the environment.

Rijkswaterstaat aims to build circularly from 2030 onward, with the purpose of having a fully circular economy by 2050. Therefore, the aim of this study is to design a reusable modular superstructure, with a small span, which can replace an existing so-called volstortligger superstructure. This is a structure that Dura Vermeer often uses and it regularly happens that these constructions are demolished after a lifespan of, for example, 10 years. This is a waste of materials and therefore an alternative will be investigated in this research. The following research question guides this study:

What is the optimal structure, in terms of shadow costs, mass and financial costs, for a modular superstructure?

In order to answer this question, a parametric study is performed using the software Grasshopper, using the Finite Element Model plug-in Karamba3D. In this way, quick structural analyses were performed, to explore multiple options. In total, more than 11.000 bridge design variants have been generated, divided over 12 different runs. The model allowed targeted optimization of shadow costs, weight and financial costs. Optimization is achieved by varying the following input parameters: number of steel girders, steel profile girders, outrigger deck, concrete deck thickness, concrete class, outrigger deck, FRP sandwich panel height, facing thickness, web thickness, and fibre volume.

In total, 6 different variants would be developed in this way, 3 with a concrete deck and 3 with an FRP deck. However, the optimization results in only 4 different variants, 3 with a concrete deck and just one with an FRP deck. The results show that the variant with the FRP deck is indicative for Shadow costs, Financial costs and weight.

After the four pre-designs had been developed, these variants were weighed against each other by means of a Trade-Off Matrix (TOM). Using this method, a distinction is made between the variants based on the pre-chosen performance indicators. All performance indicators have a weighting that adds up to a total of 100.

This study concludes that it can be advantageous to replace the traditional design with a demountable design, with the assumed boundary conditions of this study. All developed variants have an advantage with regard to environmental impact and financial costs compared to the traditional design, assuming a lifespan of 100 years in which it is moved 10 times.

FRP is increasingly utilised in the built environment, following successful implementations in the aerospace and marine industry. However, it exhibits poor stiffness and stability compared to conventional materials such as steel, making it challenging to satisfy serviceability and comfort requirements. Optimising FRP to address these challenges will allow for lightweight solutions with long service lives which could contribute to a cost-efficient reduction of carbon footprint. To this end, an optimisation tool for the preliminary design of bridges using FRP is presented in this thesis report.

The research explores the feasibility of adopting FRP in the main load-carrying system of pedestrian bridges and develops a framework for the concurrent geometry and material architectural optimisation of said structures. The study aims to achieve significant cost- and carbon footprint reductions in
monocoque FRP bridges by employing a numerical optimisation approach.

The optimisation tool utilizes the computer-aided geometric design (CAGD) software Rhino® and its parametric interface, Grasshopper®, to concurrently optimise the shape and material architecture of the bridges. Through the use of genetic algorithms, the framework overcomes FRP’s poor stiffness and
stability, and maximizes its unique advantages, including lightweight and high-strength properties, enabling free-form designs. This feat is achieved by implementing hybrid sandwich panels, comprising glass fibre-reinforced polymer (GFRP) and carbon fibre-reinforced polymer (CFRP) face sheets.
Satisfactory stiffness is ensured by defining deflection constraints, whereas constraints on the fundamental frequency and critical buckling load factor ensure adequate stability.

The research demonstrates promising results, showing potential cost reductions of up to 17% and carbon footprint reductions of up to 27.4% compared to a real case design carried out by FiReCo. However, certain limitations and areas for improvement are acknowledged, including the required run-time and the complexity of the solution space. Suggestions for enhancing the framework’s efficiency are proposed, including implementing orthotropic failure criteria and reducing the solution space through adjustments to ply thicknesses and foam core configurations.

Overall, the developed optimisation tool provides valuable insights and serves as a valuable resource for researchers and practitioners seeking sustainable and economically viable bridge designs. By embracing innovative solutions and eco-friendly materials, this study contributes to global efforts towards carbon neutrality and sustainable infrastructure development in the built environment.

Hybrid structures consisting of FRP(fiber reinforce polymer) composite decks
connected to a steel girder superstructure are gaining more attention by combining the stiffness of steel members with good fatigue endurance and high strength to weight ratio of FRP. However, when composite material FRP is exposed to elevated temperature, delamination and strength reduction could appear due to low transition temperature of fiber and resin. As a result, predicting the temperature of the FRP bridge deck becomes a preliminary requirement for the further study of the FRP bridge deck.
This research is trying to simulate the heat transfer process of the FRP bridge
deck and predict the temperature changing process when it is exposed to natural environment. To reach this goal, experiment and FEM(finite element model) are used as main methods. By detecting temperature change history of the FRP bridge deck specimen during hot weathers in Delft, Netherlands, the maximum temperature and heat transfer process of the FRP bridge deck could be studied. On the basis of experimental results, an detailed FEM in Abaqus is built up to simulate the heat transfer process of the specimen used in the experiment. After validating the accuracy of FEM result with experimental data, the FEM is used to predict the temperature of FRP bridge deck that exposed to natural environment in the hottest weather of Netherlands.
In conclusion, the results shows that the temperature of FRP structures could be
well predicted which only has 10.5% variance predicting the maximum temperature on the top surface of the specimen and high accuracy of 6% predicting the average temperature of the FRP panels along the thickness. It also has high accuracy of 3.5% predicting the maximum web-core temperature difference. The only insufficient part is that the average temperature difference of web-core has a deviation of 21% with the experimental data. As FRP panels make up the web and flange of the bridge deck which are the main structural parts that bear the stresses, the inaccuracy of predicting temperature of the core is acceptable.

The increasing traffic load on highway bridges provides a need for replacement or improvement of the bridge decks. The design of a new bridge deck using fibre-reinforced polymer (FRP) bridge decks raises questions on the connection of a FRP bridge deck to a steel superstructure. At the Tu Delft a bolted connector injected with steel reinforced resin (iSRR) is developed for the connections between steel superstructure and FRP bridge deck. Steel reinforce resin (SRR) is a combination of steel shot (SS) and resin. The steel shot is added to achieve higher stiffness and reduce the total production cost. The iSRR connection can be produced using multiple methods. For all production methods, the SRR in the connector is produced by pouring SS around the bolt followed by the injection of the resin. The iSRR connector should accompany the fatigue resistance of the FRP bridge deck. A better understanding of crack propagation can help to increase the life expectancy of iSRR connections. However, no standardized method for the testing of crack propagation in SRR exists. This research attempts to find a crack propagation testing method for SRR by investigating the use of compact tension (CT) and semi-circle bending (SCB) testing for the SRR material. For the production 5 different casting methods were compared. From these methods, the closed mould injection casting and ring mould injection casting showed the most promising production method for the production of complicated forms in SRR. From this testing the need to use modified forms of CT and SCB was found, resulting in the use of a doubly tapered compact tension specimen (2TCT) and the design of the short tapered notched beam specimen (STNB). The modified specimen forms are tested under varying loading types. The 2TCT specimen is tested under load controlled, and displacement controlled loading. From this, the incremental increasing loading type was suggested. This loading type was used in the testing of STNB specimens. Lastly with the use of FEA the stress intensity factors for the modified test forms were validated. The validation was done on the analytical formulas provided by literature for the unmodified forms. From this research, it can be concluded that the closed mould injection casting and ring mould injection both achieve good results for the production of complicated shapes with SRR. The incremental increasing displacement loading method leads to more stable crack growth. The 3 point bending testing type leads to the most promising results for crack propagation. With a linear correlation in the small range of K = 20-60 resulting in a Paris Law with 𝐶 = 4.3𝑒 − 7 and 𝑚 = 2.41. It is recommended for further crack propagation testing to produce specimens with the least amount of voids. The voids are expected to have a high contribution to the big spread of data found during the testing. For better insight, the voids in the tested specimens could also be investigated.

This research is about which geometry of FRP slabs is the most optimal, in terms of costs, to use for a given load and boundary conditions taking the occurring uplift bolt forces into account. The first part of the report focuses on the calculation of the uplift forces in the bolted connections caused by the moving loads. For this, first a detailed numerical model is made in Abaqus. Based on this most detailed numerical model, different simplifications are made to make the calculations easier and less time-consuming. Numerical models with linear bolts instead of non-linear bolts and with load superposition using a one-wheel load instead of a complete vehicle are considered. Next to this, equivalent material properties are calculated to make a model with an orthotropic deck plate instead of modeled geometry of skins and webs. For the numerical model with this orthotropic deck, the same simplifications are applied. Next to the numerical models, different analytical beam and plate models are considered. All different numerical and analytical models made, are compared on possibilities, calculation and modeling time and accuracy. Based on the comparison between the models and the capacity of the connections, it is concluded that the uplift design of the bolted connections does not need to be taken into account in the optimization of the slabs.

The second part of the graduation work is about the optimization of the FRP slabs considering the global behavior of the deck, for which a genetic algorithm is used to detect the most optimal geometry. The most important parameters for the design are the height of the deck, the spacing of the webs and the layup of the topskin, bottomskin and webs. The layup of the different elements is dependent on the number of plies, the ply orientation and the overlapping length. During the optimization process, design and initial (global) strength, stability and stiffness checks are performed. First, optimization is done for the same deck that is considered in part one of the research. Next to this case, different cases for the distance between the supports are considered for which standardization is done with respect to engineering and production of the slabs.

Based on the performed research it is concluded that the uplift bolt forces are in a range of 0.5 to 21.5 kN. Simplifications can be done by the use of an orthotropic deck with equivalent material properties. The maximum difference using this simplification is 2.9 kN, while the calculation time needed is reduced by a factor of 4-5. Other simplifications that can be done are the use of linear bolts and/or load superposition. Using linear bolts gives a maximum difference of 0.5 kN and load superposition a maximum difference of 0.3 kN, both without an improvement of the calculation time. The optimal geometry for different boundary conditions is given as a standard design with a variable number of plies for the topskin and a variable height based on intervals for the distance between the supports.

Fiber Reinforced Polymer (FRP) has increased rapidly in popularity in the past few decades. The material's advantageous properties, such as a high strength-to-weight ratio and low required maintenance, gave rise to its popularity in multiple major engineering branches. Buckling behaviour develops due to the notably low stiffness-to-strength ratio and the usual high slenderness of FRP plates. The occurrence of initial imperfections increases the tendency of the material to buckle. The load-carrying capacity of structures with post-buckling behaviour can be determined with progressive failure analysis, which requires a damage model that characterises the onset and evolution of damage. In Abaqus, the Hashin damage model is implemented by default, which considers the four failure modes of the material. Numerical analysis of strain-softening materials with local damage leads to deformation localisation in a single element: a finer mesh will decrease the amount of energy dissipated. To prevent this localisation into arbitrarily small regions, the stress is related to the deformation of a finite volume. The damage evolution is described with a stress-displacement response instead of a stress-strain response. The energy needed to open a unit area of the crack, the fracture energy, is defined as a material parameter and depends on the mesh size of the model. The assessment of fracture energy properties in composite materials is challenging due to specimen geometry and fibre lay-up, and accurate data of GFRP fracture energy is largely unknown. When no actual post-failure behaviour is acquired, the lower bound fracture energy can be determined from the material properties. Numerical analysis of uni-directional coupon experiments is performed to determine the input values and response of the lower bound fracture energy. The lower bound fracture energy implementation results in an abrupt drop in stress when the material strength is reached. Increasing the lower bound fracture energy by a minimum of 2% prevented numerical inconsistencies. Progressive failure analysis of multi-directional coupon experiments validated an increase of 10% for the lower bound values. The use of lower bound fracture energy for non-linear buckling analysis is verified with progressive failure analysis of the buckling experiments. The lower bound fracture energy, increased by 10%, approximates the ultimate strength of six tests with an average difference of 7.7%. To analyse the non-linear buckling behaviour of a GFRP plate, a buckling curve is created by varying the plate thickness. The influence of geometric imperfections on a plate's buckling strength is studied by applying different initial imperfections. Two types of boundary conditions are used to analyse if they result in a different buckling curve. A difference in the buckling strength reduction factor of 0.1 is found. Initial imperfections reduce the buckling strength of the material, which is most apparent for plate slenderness around 1.0. An initial imperfection of B/125 resulted in a 40% strength reduction compared to the elastic buckling strength. The average difference in reduction factor between an initial imperfection of B/1000 and B/125 was 16%, with a maximum difference of 26%.

Fibre reinforced polymers (FRP) can be a solution for future bridge renovation when only the deck of the bridge needs replacement. In these cases the deck is replaced with an FRP sandwich panel deck. The main advantages are a high strength-to-weight ratio which is beneficial for the ever heavier lorries and the fast installation which prevents hindrance. To connect the FRP deck with the steel superstructure, bolted connectors are used. One of the aspects that needs to be investigated before the bolted connectors can be applied are the relevant loads on the bolted connectors. The focus of this thesis will be to investigate the relevant static and fatigue forces on the connectors.
55 existing girder and arch bridges have been selected from a database of Rijkswaterstaat. For each bridge an appropriate FRP deck has been designed with an analytical method. The bridges are modelled in SOFiSTiK to investigate the shear forces in the connectors. The layout of the bolted connectors is kept constant for all bridges, as well as the transverse stiffness of the connector. The shear forces are calculated with a linear analysis. Before the static and fatigue analyses, first the hybrid interaction of the model is investigated. The hybrid interaction is the amount of horizontal forces transferred between the FRP deck and the steel super-structure. Hybrid interaction is beneficial as it increases the strength of the structure. Two models of a generic bridge have been investigated, the first is the standard model which is supposed to use hybrid interaction. The second model is an adjusted version to create a non-hybrid model. Three parameters are investigated: the deflection, the slip and the longitudinal stress over the height of the beam. The results show that hybrid interaction is created with the bolted connectors in the standard model.
In the static analyses two loads are investigated: traffic loads and temperature loads. The shear forces in the longitudinal direction are investigated as this is the governing direction. Before the loads are applied on the existing bridges, first a generic bridge has been investigated to gain knowledge over three bridge parameters. First the facing laminate of the FRP deck is changed. Second the direction of the webs of the FRP deck, and this also changes the connector layout due to feasibility. Third the expansion coefficient of the resin is changed. For traffic loads mainly the direction of the webs influences the shear forces in the connectors. For temperature loading the shear forces are largely depending on the expansion coefficient of the laminate, the close the expansion coefficient to the expansion coefficient to the steel superstructure, the lower the shear forces. The existing bridges that have been investigated resulted in a large scatter in results. The layout of the superstructure of the bridge influences the facing laminate and the number of connectors, which influences the maximum shear forces in the connectors. For both traffic and temperature loads, the connectors close to the supports experience the highest shear forces.
Besides the static loading, also fatigue is investigated. The shear forces in the connectors are calculated for one bridge, namely the approach bridge Nieuw Vossemeer. This bridge is one of the heaviest loaded bridges in the static analyses and it is according to expect judgement facing deck problems. Two aspects are investigated in the fatigue analysis, first the magnitude and second the type of load cycles. The magnitude is important as this needs to be below the slip resistance of the bolted connectors. The type of load cycles is important as this is related to how damaging the load cycle is. Because there are no S-N curves for bolted connectors between an FRP deck with a steel superstructure, the damage cannot be calculated. The R-ratio is used to investigate the type of load cycles. To calculate the R-ratio of a load cycle, the minimum shear force is divided by the maximum shear force. The resulting number expresses the type of load cycle. For the results, distinction has been made between the connectors close to the supports and connectors in the lengthwise middle of the bridge span. Both the magnitude as the type of loading is different.
Finally, it is concluded that the shear forces in the connectors is one of the aspects to be considered when designing a bridge with hybrid interaction between the FRP deck and steel superstructure. A large scatter of shear forces can be expected, depending on the bridge layout. Incorrect deck design can result in unnecessary high shear forces. The connector layout can be optimised to make the design more cost efficient.