Tez özetleri Astronomi ve Uzay Bilimleri Anabilim Dalı

Fatigue Behavior of Steel-Concrete Composite Beams Strengthened with Carbon

Fiber Reinforced Polymers (CFRP)
Bridges are vital components of highway and railway transportation systems, and determine the traffic volume of the line and the maximum load that can be carried. For smooth transportation system, it is extremely important that the bridges can be used without interruption and possess the required structural safety. There are many steel highway and railway bridges, both in Turkey and abroad. In some of these bridges, the structural safety might decrease or completely vanish in time due to a variety of factors such as fatigue, corrosion, increasing vehicle loads and traffic, additional dead loads, and environmental effects. The bridges with reduced or no structural safety should either be demolished and reconstructed, or strengthened. Generally, the costs associated with the demolition and reconstruction of structures is higher than those required for strengthening. As in many other industrializing countries, the available financial resources are limited in Turkey, and therefore the strengthening of such bridges is considered to be a better option. Moreover, the adverse environmental effects due to demolition and reconstruction should also be considered. In this context, although determining how to use the demolished material is an important problem, it should be kept in mind that natural resources are consumed to obtain new construction materials.
There are a variety of strengthening methods conventionally used for steel highway and railway bridges. However, these methods result in considerable amounts of additional dead loads to be imposed on the structures. In addition, these methods might also cause poor fatigue strength and cross-section losses. When the wide inventory of bridges in Turkey and abroad is considered, it is clearly seen that an effective, economic, and rapid strengthening method is needed for the smooth functioning of existing bridge structures. The strengthening method using carbon fiber reinforced polymers (CFRP) has come to the forefront as an alternative to the conventional strengthening methods due to increasingly falling costs, desired material properties, and easy applicability.
This study discusses the carbon fiber reinforced polymer-based (CFRP) strengthening method for the corroded and uncorroded steel-concrete composite bridge girders with determined structural deficiency and the need to be strengthened. The study focuses on investigating and improving the load-deflection and fatigue behavior of these beams under the static and dynamic (cyclic) loading. Within this framework, an experimental study was carried out, an analytical model was developed, and service conditions of bridges were examined in accordance with the specifications.
In the experimental phase of this study, along with materials tests, a total of 14 steel-concrete composite beams were tested under four-point loading. All specimens have the same scale, equal to approximately 1/6, and same fatigue sensitive details with welded stiffener connections (full-depth transverse stiffeners welded to both the flanges and web of the steel profile). Of the tests that were performed, six were static and eight were long-term dynamic (fatigue) tests. By means of the static tests, the effects of pitting corrosion damages and CFRP strengthening system on load-deflection behavior of the beams were investigated. The dynamic tests first examined the fatigue behavior of the undamaged specimens, and then the impact of pitting corrosion damages and CFRP strengthening system on this behavior.
In the analytical phase of this study, an analytical model was developed in order to predict the load-deflection and moment-curvature behavior of steel-concrete composite beams under static loads. The model includes also corrosion damages and the CFRP strengthening system. In addition, computer software was developed for easier application of the model. An existing model that was developed to determine the strength of the adhesive system between the CFRP laminate and steel beam was integrated into this software.
In the specification and design phase, the live load and fatigue load capacities of the composite beams were determined in line with the design principles of relevant specifications (AASHTO, 2002) (AASHTO LRFD, 2007) and various assumptions. The impact of pitting corrosion damages and the CFRP strengthening system on these capacity values was shown under the static and dynamic loads.
In this study, the specimens were grouped into three series:
In the first series of specimens, the fatigue behavior of steel-concrete composite beams with full-depth transverse stiffener welded connections was investigated. Corrosion damages and CFRP strengthening were not applied to these beams. A total of five specimens were tested: one static (monotonic) and four dynamic (cyclic). The load-deflection behavior was determined with the monotonic test, while fatigue behavior was determined by dynamic tests. In the fatigue tests, the specimens were subjected to cyclic loadings at different stress ranges, and stress range-cycle (S-N) data were obtained. Cyclic loadings were applied within stress ranges where fatigue fracture was expected according to the relevant specifications. In this series, no fatigue damages were observed in the specimens subjected to stress ranges correspond to less than 45% of the steel yield strength at critical fatigue points (the toe of the fillet weld connecting the stiffener to the tension flange). In the specimens in which fatigue fracture was observed, as expected, fatigue crack occurred at the bottom or edge of the bottom flange in cross-section at the toe of the fillet weld connecting the stiffener to the bottom flange. The specimens failed as this crack passed the bottom flange and progressed along the height of the web. When the results obtained from the fatigue tests were compared to the fatigue detail categories in the relevant specifications, it was seen that the specifications remained on the safe side. When the data obtained from the fatigue test are compared to the values reported in the literature related to the S-N data obtained from non-composite steel beams, it was seen that although similar results were obtained for higher stress range, higher fatigue strength was obtained for lower stress range in this study. During the cyclic loading, in general, the deflection ranges of the specimens under constant stress range increased, and therefore the stiffness of the specimens decreased. However, these changes were all very small. The change in stiffness during cyclic loading is not a sufficient parameter to determine fatigue life. The changes occurring in strain ranges at the critical fatigue points are the main determinant of fatigue behavior. The specimens with no fatigue fracture at the end of cyclic loading were loaded monotonically in order to determine the residual strength, and no significant decrease in stiffness, yield strength, and ultimate capacity were detected at the end of this loading. In addition, using the design principles in the specifications, the changes in the live load and fatigue load capacities of the specimens were determined based on the results obtained from the monotonic tests carried out after the cyclic loading. Accordingly, parallel to the decrease in the yield strength at the end of cyclic loading, the live load capacity of the specimens decreased. However, no significant change was observed in the fatigue load capacities. The comparisons were made on the basis of reference specimen with no cyclic loading applied.
In the second series of specimens, the effects of pitting corrosion damages and the CFRP strengthening system on the load-deflection behavior of the beams under the static loading were investigated. For this aim, five static tests were carried out. Four of these were monotonic and one was a cyclic (static) loading test. In the monotonic tests, corroded, uncorroded, corroded-strengthened, and uncorroded-strengthened specimens were loaded statically under four-point loading, and the impact of pitting corrosion damages and the CFRP strengthening system on the stiffness, yield strength, ultimate capacity, and load-deflection behavior were determined. These tests showed that pitting corrosion damages at the bottom flange of the steel profile resulted in a decrease in the stiffness, yield strength, and ultimate capacity of the specimen. On the other hand, strengthening using CFRP laminates helped not only to regain the stiffness and strength loss, but also to obtain a bending performance level, even higher than that of the undamaged specimen. In the beams strengthened with CFRP laminates, the yield strength and ultimate capacity increased substantially, while the stiffness increase appeared to be lower since the modulus of elasticity of the CFRP laminates used for strengthening was moderate. During the monotonic loading of the strengthened specimens, first, the steel beam yielded, then the CFRP laminates detached from the steel beam; finally the specimens failed as a result of crushing of the concrete deck slab and the buckling of the longitudinal reinforcement bars. The CFRP strengthening system increased the strength of the specimens; however, the ductility was reduced. On the other hand, pitting corrosion damages resulted in reduced both the strength and ductility of the specimens. In this study, an analytical model was developed to predict the load-deflection and moment-curvature behavior of corroded, uncorroded, corroded-strengthened, and uncorroded-strengthened steel-concrete composite beams. In order to apply the model more easily and rapidly, computer software titled “CompCurv” was developed using the “Visual Basic”. Yield strength and ultimate capacity that were predicted with the analytical model were in good agreement with the experimental study; however, the predicted stiffness values were higher than the experimental data. This is considered to be due to the assumptions of the analytical model, possible slips between the concrete deck slab and steel beam, local stress concentrations, and shrinkage cracks in the concrete deck slab. The live load and fatigue load capacities of composite beams were determined based on the monotonic tests results and the design principles in the specifications. Accordingly, corrosion damages reduced the live load and fatigue load capacities of the beams, whereas the CFRP strengthening system increased the live load and fatigue load capacities. The change in the fatigue load capacity of the beams was less than that in the live load capacity.
In the third series of specimens, the effects of pitting corrosion damages and the CFRP strengthening system on the fatigue behavior of steel-concrete composite beams with full-depth transverse stiffener welded connections were investigated. For this aim, four fatigue tests were carried out with one undamaged (uncorroded and unstrengthened) specimen, two corroded specimens with the identical corrosion damages, and one corroded-strengthened specimen. Cyclic loading range was determined as approximately 40% of the yield strength of the undamaged specimen. All specimens were loaded in the same cyclic loading range as the undamaged specimen. At the end of cyclic loading, no fatigue cracks were observed on the undamaged specimen, whereas on both specimens with pitting corrosion damages, fatigue cracks and fractures were formed. Pitting corrosion damages resulted in fatigue cracking at the fatigue sensitive details that lead to local stress concentrations. The CFRP strengthening system prevented the formation of fatigue cracks on the specimens with identical corrosion damages. Although the additional stresses occurring due to pitting corrosion damages are at tolerable levels under static loading, they caused fatigue damages under cyclic loading and reduced the service life of the bridges. The strengthening technique using CFRP laminates is a suitable method to increase the service lives of such bridges. In the specimens in which fatigue fracture was observed, the fatigue crack occurred at the bottom or edge of the bottom flange in cross-section at the toe of the fillet weld connecting the stiffener to the bottom flange. The specimens failed as this crack passed the bottom flange and progressed along the height of the web. During the cyclic loading, the increase in the deflection ranges of the specimens remained quite low for the constant loading range, and therefore no significant change was observed in stiffness. The change in stiffness during cyclic loading is not a sufficient parameter to determine fatigue life. The changes occurring in strain ranges at the critical fatigue points are quite meaningful to determine fatigue behavior. When the S-N data obtained from fatigue tests are compared to the fatigue detail categories in the relevant specifications, it was seen that the specifications remained on the safe side. The specimens with no fatigue crack at the end of cyclic loading were loaded monotonically to determine their residual strength, and the obtained results were compared to the values obtained from specimens with same characteristics (in terms of material, corrosion, and strengthening) and with no cyclic loading applied. The results showed that at the end of cyclic loading, no significant reduction was observed in the stiffness and strength features of the undamaged specimen and corroded-strengthened specimen. At the end of cyclic loading, the stiffness and strength loss on the beam strengthened using CFRP was less than those of the unstrengthened specimen. On the basis of the results obtained from the monotonic tests carried out after the cyclic loading, the live load and fatigue load capacities of specimens were determined in line with design principles in relevant specifications. These capacities were compared to the values obtained from specimens with the same characteristics and with no cyclic loading applied. Accordingly, parallel to the decrease in the yield strength at the end of cyclic loading, the live load capacity of the specimens decreased. However, no significant change was observed in the fatigue load capacities.

This thesis is composed of five chapters:

Chapter 1. Introduction: In this chapter, the subject, aim, and content of the thesis are presented and the methods used in the study are introduced. It was indicated that the specimens were divided into three series; the aim and content of the specimens in each series, the innovative features of this study and the expected contribution to the literature are explained.
Chapter 2. General Information: In this chapter the current conditions of existing steel highway and railway bridges, the reasons for strengthening these bridges (fatigue, corrosion, increasing vehicle loads, etc...) and previous studies carried out in this field are investigated. In the section related to strengthening using CFRP, information related to surface preparation, durability, adhesive systems, strengthening techniques and previous studies in this field is provided.
Chapter 3. Materials and Methods: This chapter is composed of three sections: experimental study, analytical study, and specification and design. In the experimental study section, the design of the specimens, construction of the specimens, material properties, test setup, instrumentation, and experimental program are explained. Details regarding corrosion damages and the strengthening system using CFRP are given in this section. In the analytical study section, detailed information about the assumptions, material models, equations used for modeling, behavior of adhesive, modeling of corrosion damages, and modeling of CFRP strengthening system are given. In addition the features, content, data input, use, and flowcharts of the “CompCurv” software developed for the application of the analytical model are given in this section. In the specification and design section, the design principles in the codes, loads and load combinations, cross-section and strength checks, and fatigue limit state are explained.
Chapter 4. Results: In this chapter, the results obtained from the experimental study, analytical study, and specification and design section are presented. In the experimental study, the test results of the specimens in all three series were given, and deflections and deformations occurring in the specimens are shown. In the analytical study, the specimens of each series are analyzed with the "CompCurv" software developed on the basis of the analytical model, and the results are compared to the experimentally obtained results. In the specification and design section, the design principles in the codes are explained on an exemplary specimen, and the live load and fatigue load capacities of specimens in each series are determined in accordance with the design made based on the experimental study and specifications.
Chapter 5. Discussion and Conclusion: In this chapter, general conclusions drawn at the end of this study are presented, and the results regarding experimental, analytical, and specification and design of the specimens in each series were summarized, and potential studies to be conducted in the future are mentioned.