Investigation and comparison of the seismic behavior of steel structure of moment frame with/without CFT column

- In the present research, the seismic behavior of steel structure of moment frame with CFT column has been evaluated. Finite element method has been used for numerical modeling. In the modeling, the surface and volume elements were used for steel and concrete, respectively. One, two, and three-story frames with and without CFT column under different earthquakes were investigated through dynamic analysis of the time history. The results show that the steel frame with CFT column, in terms of energy dissipation, has better behavior than the steel frame. The roof displacement in the different frames with CFT is less than that of the steel frame which indicate the improved performance of the frame. Using the CFT column in the column, in comparison with steel column, has led an increase in the bearing capacity of the structure under the earthquake. By adding the concrete to the column, the failure status of the structure will change. In the model with CFT, the failure occurs mainly in the location of the panel zone, while in the steel column under earthquake, the buckling occurs in the column. The reason for that may be due to steel-concrete interaction, which in turn, will increase the bearing capacity of the CFT column. Presence of CFT column in the steel frames will result in increased bearing capacity of the frame. Filling the steel frame with concrete prevents the inward buckling of the column and delays the local buckling of the steel wall.

The extent of elastic stiffness of CFDST columns confined with more rings is not higher than the non-confined CFDST columns [9]. Bridge and Oshea (2000) concluded that by increasing the eccentricity force, will improve the ductility of short columns. In the concrete-filled columns, an important effect of concrete is that the local buckling of steel wall is delayed and the concrete in confinement mode is capable of bears higher stresses and strains than the non-confined case. The advantages of the double-layered composite columns compare with the single-layered composite columns include less weight, higher ductility and bearing higher axial forces. The scenario which occurs in the most scientific problems in a reinforced concrete compression member is the coexistence of axial load and bending moment. In the reinforced concrete frames, by cast-in-place concrete, even if the openings of both sides of a column are same in size, because of variations of live load, the bending moment will occur in a column. Even in the case of using preconstructed columns, the vertical loads result from falling and rising the beams, a moment will occur in the column. Based on these facts, even in the cases wherein, theoretically, a pure axial load will apply on a reinforced concrete column, the regulations recommended that applying a bending moment is necessary so that an eccentricity besides the axial load [10]. Talebi et al. (2018), investigated a 3D nonlinear finite element (FA) model to study the response of concrete-filled tube columns (CFT) under the fires after an earthquake. In their modeling, they considered three consecutive steps include cyclic, thermal, and structural analyses. Three possible seismic damage scenarios have been considered. They have considered high damage level and assumed that while the column has maintained its overall stability after the earthquake, has reached 50% its lateral strength. The results showed that the damages of the upper area have no effect on the fire response of the damaged column. Besides, the column with a damaged middle opening has less strength under fire, due to concurrency of damage location. Hanifehzadeh and Mousavi (2019), simulated the response of an SCS wall exposed to the near-field explosion load. The structural performance of a regular SCS panel which exposed to the near-field explosion load has been evaluated in terms of maximum damage and deformation. The maximum deformation, plastic strain, and damage mode have been investigated in different loading scenarios. The purpose of their study was to predict the structural reaction on SCS panel exposed to different explosions and identifying the optimal configuration in terms of concrete strength and plane thickness. The optimal parameter for each of the structural components is identified using a specific optimization method. It was concluded that using the proposed wall configuration compares with the regular walls will have better performance while its additional construction costs is insignificant [12]. Yang et al. (2019), carried out a case study of 491 test results of the CFT under axial load from the available literature. The obtained results were used to investigate the effect of loading conditions, diameter to width ratio, concrete type, and compression strength on the bearing capacity of CFT columns wherein the models were used for the stress of steel tubes through nonlinear regression. By using the analysis based on Von-Mises performance along with the proposed models for ring stress, a model was proposed for predicting the bearing capacity of CFT columns. The performance of the proposed model and the available model was investigated using the data bank. Since the model development trend was rational, the proposed model provided a more reasonable and accurate of bearing capacity of CFT columns in different loading conditions [13]. Xian et al. (2019) proposed a new type of confined base-column connections using metal packaging for limiting the external elements of the reinforced concrete and investigated the seismic behavior of both traditional and proposed (CFT) concrete column-base connections experimentally by conducting 8 tests. The samples were simulated in large scale under seismic loads. The tests parameters were mainly including thickness, height and flexural strength ratio of external elements with or without metal containers and shear nails. The effects of test parameters on the failure mode, response curves to load deformation, stiffness, strength, flexibility, ability to dissipate energy and distributing the strain of confined beam-base connections were investigated comprehensively. The results of the tests showed that with the proper design of beam-base connections, desire seismic behavior can be achieved and by using the metal packaging, the cracking of external parts of reinforced concrete can be prevented effectively [14]. Shariati et al (2020) carried out a numerical investigation on the behavior of concrete-filled tube columns under the axial pressure. They created a finite element (FE) model to simulate the behavior of CFT columns. By comparing the FE results and the tests, it was found that the numerical model can capture the desired conditions and can precisely predict the axial performance of the CFT columns. Also, with an increase in the thickness of the steel tubes, due to the higher effect, the bearing capacity of the columns has increased. By increasing the strength of concrete, with more cross-section, load-bearing capacity has increased compared with increasing the steel tube. In CFT columns with more cross-section, the strength of concrete is more than the load-bearing capacity which is more considerable in columns with welding point [15]. By studying the previous literature it can be concluded that the behavior of steel structures with moment frame with and without CFTCFT has not been yet investigated and needs to be investigated and studied.

Methodology
To create the base model, ABAQUS software was used. For steel column and the concrete core, the surface and volumetric elements were used, which were modeled with the reduced integral equation with solid and brick elements with 3 degrees of freedom for volumetric model and S4R for the surface model. The reduced integration elements for meshing the model. Loading has been applied uniformly on the upper plane so that the load is applied in the opposite direction and as similar form with the cyclic loading protocol. The displacement control load is applied to the model. The upper and lower planes have been connected to the desired column via Tie command. In this research, the implicit dynamic analysis has been used. the concrete-steel interaction has been taken in to account. The contact algorithm was used to define the concrete-steel contact. The behavior of hard contact and penalty algorithm were used. The finite element analysis of CFT frame provides a better understanding of real behavior with minimum cost in minimum time. The test base model shown in Fig. (1) was used for modeling.

Materials' specifications
The materials were modeled nonlinearly in the modeling software. The nonlinear behavior with plasticity behavior was considered for steel. For the plasticity model of concrete, the concrete damage plasticity model was used.

Loading and boundary conditions
The support conditions were specified for the model. Fig. (5) shows the model of the concrete-filled column with the support conditions. The test conditions of the sample on the jack force were simulated. To do this, the sample was located on the frame under the displacement according to the SAC protocol. However, to apply the earthquake, the earthquake displacement was applied to the location of the columns' support. Applying the load to the structure was done as the drift control. The SAC loading protocol was used. The cyclic loading pattern is shown in Fig. (6).

Research results
In this section, different samples have been investigated. For better comparison, the 1, 2, and 3-stories steel moment frames and concrete-filled columns were compared. First, the loading of frames was periodically and as displacement control and the results for studied frames were compared. Then, the studied samples were exposed to the displacement of some real earthquakes and the time-displacement and hysteresis curves of the samples were compared. The results are discussed in the following.

Investigation of hysteresis behavior of models
Hysteresis curves of two frames with and without concrete-filled columns have been compared. Fig. (7) and Fig.  (8) show the energy dissipation curve of steel frame and energy dissipation of concrete-filled steel frame, respectively. As can be seen from these figures, the frame with reinforced concrete-filled steel column has higher energy dissipation and less strength loss. The behavior of steel moment frame with concrete-filled steel column is better in terms of energy dissipation.

Investigation of the seismic behavior od models
Deformation and stress of samples without concrete-filled steel column are shown in Figs. (9)- (12). As can be seen from these figures, the nonlinear behavior with high energy dissipation level has occurred in the samples. The behavior in the concrete-filled sample is better since the surface confined to the residual rings in higher which indicates the higher energy dissipation in the frame with concrete-filled column. In Fig. (10), the column buckling in the near-earth location is entirely clear.   As it is cleared in the figures, the presence of concrete in the steel column has led to a reduction in stress level and also an increase in the column capacity against the buckling. This is more obvious when the beam behavior became nonlinear in the beamcolumn connection location. Indeed, the presence of concrete in the steel column has concentrated the failure to the panel zone and near the beam-column connection, which indicates the improved frame behavior due to the presence of concrete.

Conclusion
In the present study, the investigation and comparison of the seismic behavior of steel structures of moment frame with and without CFT column were addressed. To compare the bearing capacity and the seismic behavior of structures, the 1, 2, and 3-stories models under earthquake were investigated. By investigating and comparing the displacement time history of models with and without concrete-filled columns, the following results are obtained:  The concrete-filled steel frame has better behavior than the steel frame in terms of energy dissipation.  Displacement of different concrete-filled frames is less than the roof displacement of steel frames which indicates the better performance of these frames.  Presence of concrete in the steel frame has led an increase in the bearing capacity of the column under the earthquake and periodic load.  Presence of concrete can affect the frame behavior so that the failure in the concrete-filled column occurs in the panel zone and the column has usually appropriate behavior, while in the steel frame under earthquake the buckling occurs.  Contact and interaction of concrete and steel have led an increase in the bearing capacity of concretefilled steel column.  In the concrete frame, the presence of concrete-filled column has led an increased load-bearing of the frame.  Filling the steel column with concrete has prevented the inward buckling of concrete and has delayed the local buckling of the steel wall.