Visualized Simulation And Safety Assessment For Progressive Collapse Of RC Space Frames With And Without Infilled Walls

It has passed for nearly11years since the event "911" in USA occurred,however, the collapse scene of the World Trade Centers (WTC) has been alwayswarning the people that the progressive collapse of building structures will result insignificant casualties and property losses. Since that catastrophic event, the problemof progressive collapse prevention design and robustness assessment for engineeringstructures has been paid more and more attention to in the community of civilengineering all over the world. In China, the research on progressive collapse ofstructures was started later than other developed countries. Specifically, there haslittle research on progressive collapse of RC space frames with infilled walls. In thisthesis, OpenSees is taken as a platform for finite element analysis, and DONAP isused as a tool for visualized simulation. On the basis of simulation anlaysis andverification by experimental results, the problem of safety assessment for progressivecollapse of RC space frames with infilled walls is systematically and thoroughlystudied. The main contents of this thesis are as follows:(1) The software OpenSees has drawn much attention of researchers worldwidebecause of its high precision in nonlinear numerical simulation. However, thevisualization is it’s a weak point for further applicatoins. To realize the visualizationof nonlinear static and dynamic response as well as progressive collapse structures,the author develops a graphic user interface for OpenSees, named as OpenSeesGUI,in visualized simulation software DONAP, by co-working with the graduate studentsfrom the School of Software in Tsinghua University. In OpenSeesGUI, the animationwhich simultaneously follows with the dynamic results is realized, and the problemof the unsteady running of OpenSees in DONAP is also solved.(2) To verify the accuracy of the interaction model of fiber sections of infilledwall, a group of tested RC frames with infilled walls are selected as case studyexamples. These infilled frames are modeled by OpenSees, and the results from finiteelement analysis are compared with the ones from experiments. It is shown that theP-delta hysteretic curves and skeleton curves from the simulation and theexperiments have large deviations in the range of plastic deformation, the reason ofwhich is that the ideal elasto-plastic material adopted by the infilled wall model doesnot conform with the practical materials of the infilled walls. Therefore, in this thesis,a hysteretic material model instead of the ideal elasto-plastic material is introduced inthe original infilled wall model, the simulation results demonstrate that the improvedmodel can fit the tested results very well.(3) To investigate the progressive collapse resistance of RC space frames withinfilled walls，the OpenSees is utilized to build the analysis model of a two-bay and six-storey RC space frame with infilled walls. Two columns of the model on the firstfloor are removed instantaneously. The information of the structural responses isobtained by nonlinear dynamic analysis (NDA), and is compared with the resultsfrom experiments to validate the accuracy of simulation. The fiber-section interactivemodel is taken to simulate the infilled walls in the structure. The results show thatnumerical simulation resluts are in accordence with the experiment ones. Thestructure does not collapse in the case of removing two columns of the first floor. Theprogressive collapse resistance of RC space frames with infilled walls is larger thanthe bare frame without infilled walls. The infilled walls can greatly improve theprogressive collapse resistance of RC space frames.(4) To quantitatively assess the progressive collapse resistance of RC spaceframes with and without infilled walls, static pushdown analysis (PDA) and VerticalIncremental Dynamic Analysis (IDA) methods are utilized. Based on the progressivecollapse pushdown curves and IDA curves, three robustness indices, namely, strengthredundancy factor (SRF), residual influence factor (RIF), and damage strength ratio(DSR), are calculated for the RC space frames with infilled walls. It is found from theanalysis results that the values of these indices from PDA are bigger than those fromVertical IDA. Therefore, for the frames with infilled walls, the method of PDA ismore conservative than the vertical IDA. It is also found that the robustness indicesof the frames with infilled walls are larger than the ones without infilled walls forboth the PDA method and the Vertical IDA method, the main reason of which is thatthe infilled walls in structures can improve the robustness of RC space frames.(5) To investigate the fragility of RC space frames with and wihout infilledwalls in the case of removing two columns of the first floor, the uncertainties existingin the structures are considered, and one hundred samples are derived based on animproved Latin hypercube sampling (LHS) method. Both static PDA and vertical IDAmethods are utilized to obtain the progressive collapse fragility curves for onehundred two-bay and six-storey RC space frame buildings with and without infilledwalls respectively. Meanwhile, probabilistic properties of the design parameters andload combinations are considered, the computational non-convergence is taken as acriteria of progressive collapse of structures. The results show that the risks ofprogressive collapse both the frames with and without infilled walls are limited in thecase of removing two side columns on the first floor; and that the progressivecollapse resistance of RC space frame buildings with infilled walls is larger than theones without infilled walls.