Progressive Collapse Resisting Capacity of Moment Frames with Infill Steel Panels

Transcription

1 Paper 23 Progressive Collapse Resisting Capacity of Moment Frames with Infill Steel Panels M.H. Lee, H. Lee and J.K. Kim Department of Architectural Engineering Sungkyunkwan University, Suwon, Korea Civil-Comp Press, 2012 Proceedings of the Eighth International Conference on Engineering Computational Technology, B.H.V. Topping, (Editor), Civil-Comp Press, Stirlingshire, Scotland Abstract In this paper the progressive collapse behaviour of a moment frame with infill steel panels is evaluated using nonlinear static pushdown analysis. The analysis model is a two storey two span structure designed only for gravity load, and the loaddisplacement relationship is obtained with the centre column removed. To obtain local stress and strain as well as the global structural behaviour, a finite element analysis is conducted using ABACUS. Through the analysis the effect of the span length and the thickness of the steel plate on the progressive collapse behaviour of the structure is investigated, and the effect of the dividing the infill panel using stud columns is also studied. According to the analysis results, the thickness of the panels required to prevent progressive collapse increases as the span length increases, and as the number of panel division increases the progressive collapse resisting capacity increases slightly but the effect is not significant. Keywords: special truss moment frames, progressive collapse, nonlinear analysis, energy based design. 1 Introduction Infill steel plates consist of vertical steel plate infills connected to the surrounding beams and columns and installed in one or more bays along the full height of the structure mainly to resist lateral load. According to previous research, they exhibit high initial stiffness and dissipates significant amounts of energy. Timler and Kulak [1] found that the steel plate wall with unstiffened thin plates had high ductility as well as high strength even after the local buckling of the thin infill plate. Driver et al. [2] found that the specimen showed a ductile behaviour without brittle failure at the connection. Park et al. [3] found that to achieve large ductility, the boundary columns must resist the combined axial force and transverse force developed by the tension field action of the infill plates. Formisano et al. [4] investigated the use of 1

2 steel and aluminium shear panels as seismic retrofitting systems of existing RC structures. They concluded that the thin plates could be considered as effective strengthening devices of existing RC framed structures. In this study the effect of infill steel panels on enhancing the progressive collapse resisting capacity of steel moment frames was investigated. To this end two story two bay model structures were designed with only gravity load and were retrofitted with steel panels located in the second story. The progressive collapse was initiated by arbitrary removing of the first story center column. The progressive collapse resisting capacity of the model structures was investigated both by nonlinear static and nonlinear dynamic analyses using the finite element program code ABACUS [5]. 2 Analysis modeling A two-story steel moment frame was designed as a prototype structure. The model structure has uniform story height of 3.6m and the span length is varied from 6 to 12m. Only the two dimensional exterior frame was analyzed for progressive collapse. Fig. 1 depicts the plan shape and elevation view of the analysis model structures with 9m span length.table 1 shows the member sizes of the analysis model structures. The columns and beams were designed using SM490 and SS400 steel, respectively. The steel plate used for retrofit was made of SS400. 2@9 m 2@9 m (a) Plan of 2-story model (b) Elevation of 2-story model Figure 1. Structural plan and elevation of analysis model structures Column Girder 6m span 250X250X9/14 400X200X8/13 9m span 400X400X13/21 582X300X12/17 12m span 550X550X20/28 650X420X20/28 Table 1. Member size of model structures (mm) 2

3 Fig. 2 shows the simplified stress-strain relationships of steels obtained based on experimental results [3]. The Young s modulus is MPa and Poisson s ratio 0.3. The beams and columns were modeled with eight node solid elements and the infill steel panels were modeled with four node shell elements. 600 Stress(MPa) Columns Girders, Plates Strain Figure 2. Stress-strain relationship for structural steel Four types of infill steel panels were applied for analysis: (i) a full steel plate within a span; (ii) a partial steel panel located between stud columns; and (iii) a partial steel plate with 1/3 of story height. The configuration of the model structures are shown in Fig. 3. The partial steel panels shown in Figs. 3(b) and 3(c) may be applied to accommodate openings such as doors and windows, respectively. The steel panels with holes may be used to save steel tonnage or to provide openings. (a) Full infill plates (b) Partial infill plates at center (c) Partial infill plates at sides (d) Infill plates with 1/3 of story height Figure 3. Configuration of infill steel panels used in the analysis For simulation of progressive collapse, nonlinear static pushdown analyses were carried out by removing the first story center column and gradually increasing the vertical displacement in the location of the removed column. For nonlinear static analysis the amplified gravity load of 2(Dead load Live load) was applied 3

4 following the recommendations of the GSA guidelines [6]. For dynamic analysis the unamplified load of (Dead load Live load) was applied. 3 Analysis results Pushdown analyses of the model structures were carried out using the program code ABACUS [5] to investigate the progressive collapse resisting capacity and the failure mode when a first story column was removed. Figs. 4 depicts the pushdown curves of the model structure with 9m span length installed with full infill plates with various thickness. The first story middle column was arbitrarily removed and the applied force, normalized by the imposed gravity load specified in the Guidelines, vs. the vertical displacement relationship was plotted. 5 4 Plate 8mm Plate 6mm Plate 4mm Load Factor Plate 2mm Bare Frame Vertical Displacement (mm) Figure 4. Pushdown curves of the 9m span model structure with full infill panels The pushdown curve of the bare frame without steel panel was also obtained for comparison. It can be observed that the maximum load factor of the bare frame fails to reach 1.0, which implies that the structures lack enough strength to resist the specified load. When full steel panels were installed, the maximum strength exceeded the load factor of 1.0, implying that the structure can resist the load combination of the GSA guidelines [6]. The strength kept increasing as the thickness increased up to 8mm; however to add 2mm thick steel panels would be enough to resist the progressive collapse of the structure subjected to sudden loss of the center column. Fig. 5 shows the stress distribution in the steel panels when the vertical deflection at the top of the removed first story column reached 5cm. The formation of tension field is clearly visible in the 2mm thick plate, and the stress concentration reduces gradually as the thickness increases. 4

5 (a) t=2mm (b) t=4mm (c) T=6mm Figure 5. Stress distribution in the infill steel plate when the vertical displacement reached 50mm 4 Full Plate Plate at Side 3 Load Factor Plate at Center No Plate Vertical Displacement (mm) Figure 6. Nonlinear static analysis results of the 9m span structure with 4mm thick partial infill walls 5

6 The pushdown results of the 9m span structure with 4mm thick steel plate located only at part of the span as shown in Figs. 3(b) and 3(c) are depicted in Figure 6. It can be observed that the strength of the structure with partial steel plates with only 1/3 of the full width at the center of the span is reduced significantly compared with that of the structure with full width infill plates. However the reduced strength is still almost twice that of the bare frame and is enough to resist the progressive collapse. When two partial plates were installed at both sides of a span, the strength increased slightly; however considering the doubled amount of steel plate, the increase in strength is only minute. Fig. 7 shows the stress distribution in the partial steel plates when the vertical deflection reached 10cm. It can be observed that the partial infill plate placed at the center of the span is more highly stressed than those placed at the sides of the span when the structure is subjected to the same vertical deflection. In addition to the possibility of providing opening, the partial steel infill wall may be effective in preventing progressive collapse with reduced steel. Figure 7. Distribution of stress at partial steel plates when vertical displacement reached 10cm Fig. 8 compares the pushdown curve of the 9m span structure with partial infill plates with 1/3 of story height (h=1.2m) with those of the structure installed with full infill plates. The added horizontal beams are H-shaped sections with the size of

7 Fig. 8 Nonlinear static analysis results of the frame installed with partial steel plates with 1/3 of story height 5 Summary Infill steel plates consist of vertical steel plate infills connected to the surrounding beams and columns and installed in one or more bays along the full height of the structure mainly to resist lateral load. According to previous research, they exhibit high initial stiffness and dissipate significant amounts of energy. In this study the effect of infill steel panels on enhancing progressive collapse resisting capacity of moment frames was evaluated. The progressive collapse potential of the analysis models was evaluated based on an arbitrary column removal scenario. Nonlinear static and dynamic analyses of the model structures were carried out using the finite element analysis code ABACUS. The performance of the structures with infill panels divided by stud columns was also studied. The analysis results showed that the infill steel panels were effective in reducing the progressive collapse potential of moment frames. It was observed that as the thickness of the steel panel increased the progressive collapse resisting capacity also increased. However when the thickness of the steel panels increased higher than a certain level the increase in the progressive collapse resisting capacity is not significant because columns yielded prior to the yielding of steel panels. Even when the infill panels were installed in only a part of the span the progressive collapse resisting capacity was somewhat increased. The simplified single brace model of an infill panel corresponded well with the finite element model. 7

8 Acknowledgement This research was financially supported by a grant (Code# 09 R&D A01) funded by the Ministry of Land, Transport and Maritime Affairs of Korean government. References [1] Timler, P.A., Kulak,G.L. (1983) Experimental study of steel plate shear walls, Structural Engineering Report, 114, University of Alberta, [2] Driver, R. G., Kulak, G. L., Kennedy, D. J. L., and Elwi, A. E. _1998_. Cyclic test of a four-story steel plate shear wall. J. Struct. Eng.,124(2), [3] Park, H.G., Kwack, J. H., Jeon, S.W., Kim, W. G., and Choi, I. R. Framed steel plate wall behavior under cyclic lateral loading. J. Struct. Eng.,133(3), (2007), [4] Formisano, A., De Matteis, G., Mazzolani, F.M., Numerical and experimental behavior of a full-scale RC structure upgraded with steel and aluminium shear panels, Computers and Structures 88 (2010), [5] ABAQUS CAE (2007) General Finite Element Analysis System for Windows. [6] GSA Progressive Collapse Analysis and Design Guidelines for New Federal Office Buildings and Major Modernization Project, The U.S. General Services Administration. 8