| Concrete filled steel tube columns are now widely used in the construction of many tall buildings. This infatuation for this kind of structure is explained by the fact that it combines the advantages of both steel and concrete. One of the main benefits when it is compared with bare steel tubular column is a substantial increase in load bearing capacity of the column due to the concrete filling. In addition, a higher fire resistance can be obtained, this is most likely because the concrete core acts as a heat sink and the steel section heats up more slowly and the loading is transferred to the concrete core. Concrete-filled steel tubes also have much better endurance characteristics than conventional reinforced concrete columns under fire conditions as the steel cage prevents spalling of the concrete, which remains better protected against fire. Another benefit of the use of such a structure is that the tubular form of the steel eliminates the need for formwork for the concrete.Until not so long ago, the elimination of fire protection from this type of construction was supposed to be an economic consideration; however, it has now become an important safety issue after the September 11th terrorist atrocity on the World Trade Center in New York in 2001. Thus, fire protection effectiveness has become another key point for a structure which has become often used.Growing concerns about the safety of structures in general and under elevated temperature conditions have been witnessed during the last few decades. Indeed, concrete and structural steel constructions are exposed to elevated temperature in many types of structural loading. Nuclear power stations and accidental fires in residential buildings are the most evident examples of such structures. The theory of the behavior of the structure in the event of a fire is still not fully developed due to the complexity of the problem and the large number of parameters controlling this behavior. The prediction of the fire resistance of a steel and concrete construction under accidental fire is important for the design of the structure so that it resists load for a period of time sufficient for the rescue operations. It is also important for subsequent assessment of the residual capacity of the members of the structure for its rehabilitation in the aftermath of the fire.Thus, analysis of the structural behavior of the structure in relation with fire is an important part of the design for fire safety of steel structures. But due to its complexity, it is not possible to estimate the global structural behavior by the use of manual calculations. On the other hand, laboratory tests based on the standard fire test have practical and economical limitations. As a consequence, the development of numerical models, usually based on the finite element method, has been use to try to answer to this problem. The behavior of concrete filled steel tube columns has been greatly simulated during fire but on the behavior after exposure to fire, fewer tests have been carried out.For concrete filled steel tube members, the calculation of fire resistance involves the determination of temperature distribution, deformation of the elements under temperature expansion and compression, and the strength reduction. All these qualities vary with time, and the calculation is a dynamical process very complex. Development of numerical methods, such as finite elements, and using a high-speed computer, can give great benefits in solving this problem. The use of this method can also enable an analysis of the behavior of some specific areas of the structure, such as the interaction between steel and concrete, which has mainly been studied at ambient temperature.The calculation of fire resistance can be performed by the following steps. The first step is to calculate the temperature distribution. Unlike in the steel structures, the temperature of different locations in the concrete filled steel tube section is diverse and in which the properties of concrete and steel vary with the fire. The next step is to calculate the deformation and stress in the member in continuous time. The mechanical properties used for the thermal stress analysis are renewed at each time step since its temperature is of a time-dependant character. A considerable amount of data on the thermal and mechanical properties of various materials such as steel and concrete, have become available over the last decade, which improve the analysis. However, the studies made so far have only been focused on some stage of what a concrete filled steel tubular column is supposed to be confronted to, when it is exposed to fire while being a member of a whole structure. Indeed, few considerations have been made concerning the possible influence of the loading and heating history on the post fire behaviour of such column.This research proposes to analyze the behavior of concrete-filled steel tubular stub columns under heating and loading using finite element methods and the program ABAQUS. The study is focused on the whole process that is applied on the columns, which includes the loading of the element at a constant initial temperature, the exposure to high temperature and the cooling down period during constant loading and the post fire loading at constant initial temperature until rupture. The main points of this analysis are to take into consideration the history of the constraints to which the column is exposed, comprising load and temperature, and the interaction between steel and concrete, both of which have been neglected in previous studies.The aim of the research is to obtain a model able to predict the structural behavior of the concrete filled steel tube stub columns during the experiment and to gain a better understanding of the influences of the different parameters, such as the steel ratio, the steel yield strength, concrete compressive strength, the fire time duration or the initial load ratio, on the behavior of the element.In order to achieve this goal, the study has been divided into several steps. The separations of the study in those different parts are caused by the nature of the material properties involved in the experiment. Since the properties of steel and concrete are hugely dependant of the material temperature, the first stage is used to calculate the temperature field in the section of the concrete filled steel tube column. To fulfill this task, the program ABAQUS is used to build a finite element model which is confronted to experimental data collected in tests. Once the model predictions show good agreement with the experimental records, it can be used to calculate the temperature in the column section and thus used to create a model to compute the materials behavior at each time steps of the study. The second step is the creation of an ABAQUS model analyzing the strain-stress behavior of the column. This stage needs to consider the interaction between the steel tube and the concrete core and the differences of the materials behavior between the increase of temperature period, the cooling down period and the post-fire loading. The tests data collected during experiments by other researchers will enable to assess the model which in return can be used to analyze the influences of the different parameters on the behavior of a column exposed to such conditions. This understanding can be of a great importance in the design of future columns or in effectiveness of possible repair for damaged elements.The model used in this study, and created by using the program ABAQUS, to analyze the load-deformation of the concrete filled steel tubular stub columns, is dependant of the temperature field analysis. The heat transfer analysis results acquired will thus be employed as input in this part. The use of data resulting from a previous analysis implies a need of high coherence and compatibility between the two models. For this purpose, some assumptions have been established, while the hypotheses that have been made to simplify the heat transfer problem are still valid in this analysis. The latter were that in the column the temperature does not vary in the longitudinal direction so that the temperature field can be considered as a 2-dimensional temperature field and that it is supposed free of internal heat source. In addition of these suppositions, the concrete filled steel tubular stub column will be considered to be ended on both sides of a thin plate made of a very stiff material on which the loads will be applied. This element was needed to simulate the uniform repartition of the load on the column section.Furthermore, numerous parameters were defined to establish the boundary conditions, materials properties and element divisions which were needed in the different analysis.The aim of this analysis is to predict the post fire mechanical behavior of the specimen tested while taking into account its heating and loading history. Thus, the model needs to be able to conduct successively load-deformation analysis under normal and high temperature, after exposure to high temperature or during the cooling down period. Each of these stages is associated with a specific stress-strain relationship of steel and concrete. The models used to calculate the materials mechanical properties, depending on the temperature and the conditions involved, exploit formulae that have been established by using test data.The boundary condition is an important issue in this study since it defines the stimuli to which the structure has to respond. The whole problem is based on the influence of the fire temperature and the loading on the element, a large number of different heating and loading history has therefore been considered.The loading carried out on the column is considered to be applied on the bearing plate, consideration which allows a uniform repartition of the load on the CFST stub column section, and the mechanism of heat transfer includes the radiation and convection effects The sequence of heating and loading to which the columns are exposed are decomposed in four categories. The first one is associated to columns exposed to a high temperature TF which has been so slowly increased that the temperature in the column section is uniform and identical to the fire temperature. Once the temperature is TF reached, the column is loaded. The second category is composed of columns which have been exposed to high temperature TF long enough so that the temperature in the section is uniform and identical to the fire temperature. The column is then let to cool down long enough for the temperature in its section to be uniform and equals to the ambient temperature, and finally loaded. The third is represented by an exposition to ISO-834 standard fire and its cooling down stage, which create non-uniform temperature fields in the column section, followed by its loading. The last category is composed of the sequences containing a loading stage at constant temperature T0 which load the specimen up to a load NF, a heating stage following the ISO-834 standard fire curve at constant load NF, a cooling stage at constant load NF and a loading at constant temperature T0.The problem posed in the attempt of calculating the load deformation of the concrete filled steel tubular stub column presents, according to the hypotheses formulated, some rather interesting geometrical symmetry in its conception. This observation can be used to lighten the meshing and the memory consumption in the analyses of the element studied. Indeed, the whole column does not need to be taken into account during the analysis since a quarter of it respects the demands of the problem and some arrangement with the loading intensity can enable this modification. In the following analyses, only a part of the concrete filled steel tubular column will be then designed to supply the results of the study. For each symmetry plane defined, specific boundary conditions have to be delimited to limit the deformations along those surfaces.During the creation of the model, the concrete core and the bearing plate were built as solids while the steel was build as a shell. For computational accuracy, 9 points Simpson integration points were used in the shell element thickness. The interaction between steel and concrete during the analysis was defined by a surface-to-surface contact in the interaction manager. Steel was considered as the master-surface and concrete as the slave-surface while the contact interaction property allows a hard contact in the normal behavior and define the friction coefficient in the friction part of the tangential behavior category.From the models created in this research, three major points can be reminded.(1) The finite element model created to predict the temperature field in the section of the column has been confronted with experimental tests results. Those tests displayed different specimens for which the diameter dimension, steel wall thickness, fire protection coat thickness, time of fire exposure or fire temperature curve varied. The comparisons between predictions and experiments showed good agreements, implying a good reliability of the heat transfer analysis of the model. Moreover A parametrical analysis concerning the section diameter, steel wall and protection coat thicknesses has been exposed, which showed the relatively huge impacts of the different characteristics. An analysis on the influences of the concrete core on the heating and cooling process of the concrete filled steel tube column has also been presented, displaying its heat regulator capacity.(2) The finite element model created to predict the stress strain relationship of the CFST column during the different stage of the process has been confronted with experimental tests results. Those tests included analysis under normal or constant high temperature, after exposure to constant high temperature or after exposure to ISO-834 standard fire curve. The comparisons between predictions and experiment results showed good agreements, implying a good reliability of the mechanical analysis of the model.(3) The analysis on the behaviour of the concrete filled steel tubular stub column has been completed through the use of the finite element model and a succession of different stage including loading under constant temperature, heating and cooling under constant load, and loading again under constant temperature. The results drawn from the parametrical analysis show that the initial load value, time of fire exposure and steel ratio have an important impact on the residual strength index which increase with the growth of those parameters. However, the steel yield strength and concrete compressive strength do not have a huge influence on the residual strength of the column.During the parametrical analysis made in this study, the following results can be extracted.(1) With an initial load ratio comprise between 0.4 and 0.8, the influence of the initial load value on the residual strength emphasizes on the importance of the steel concrete interaction and the confinement effect. Indeed, during the heating, the higher the initial load is, the stronger the interaction between steel and concrete will be and the better their cumulated action will be in the post fire stage, increasing the residual strength of the column.(2) The steel ratio appears to be the characteristic of the CFST stub column that have an impact on the residual strength, and the fact that it increases with the steel ratio can be imputed to the proportion of steel present in the column, steel which will regain some of its performance during the cooling stage, thus increasing residual strength of the column.(3) The variations of the residual strength index that have been observed during the study, attest that the residual strength of the column can be different with its ultimate strength obtained under normal temperature without loading. However, the axial deformations associated are important, and have to be taken into consideration when the workability of the structure is regarded. Finally, it can be pinpointed that no test record concerning this experimental process exists, which prevent to assess the model in its entire evolution. Furthermore, the finite element model has been verified in a segmented way, however, the suppositions concerning the concrete properties during the cooling stage, which would be those of the post fire stage, have to be confirmed by some experiment, actually inexistent. Further work is needed in these two areas. Moreover, the present study was focused on the concrete filled steel tubular stub columns, more researches have to be done concerning longer column, beam column or specimen with different geometrical section for a better understanding of this phenomenon.
|
PDF Full Text Request