Cellular Automata-based Prediction Of Earthquake Failure Mode Of Cold-formed Steel Framed Composite Shear Wall Structure

The cold-formed steel(CFS)framed composite shear wall structure has been widely used in low-rise residential buildings in Europe,America,Australia and Japan,because it possesses the characteristics of light weight,high strength,environmental protection,energy saving and quick construction.The development of the traditional CFS framed composite shear wall structure from low-rise to mid-rise conforms to Chinese basic national condition of more people with less land,and it is also an effective way to speed up the development of housing industrialization and prefabricated assembly.In the process of promotion and application of the CFS framed composite shear wall structure,the research on seismic performance and failure mode of the whole structure is a key issue.As a new method for research,the cellular automata(CA)method had been successfully applied to investigate the failure mode of the reticulated shell and can better predict the failure mode of the reticulated shell under the earthquake.Therefore,the CA method is introduced into the prediction of the failure mode of the regular mid-rise CFS framed composite shear wall structure in the thesis,and a new analytical idea is further provided for seismic performance of the overall light steel structure of this kind.Firstly,the unloading stiffness in the simplified model of the CFS shear wall proposed by the research group of Prof.Ye did not match ideally with the experimental results.To figure out the above problem,a modified material constitutive model was proposed according to the characteristics of the load-displacement curve of the wall under the reciprocating load,and the add-on development of OpenSEES was carried out in material property module.The modified material model was applied to define the diagonal support of the simplified equivalent wall support model.The simulation results showed that the nonlinearity,strength degradation,stiffness degradation as well as pinching characteristics of the simplified analysis model of the wall were basically consistent with the experimental results.Secondly,a simplified model of the mid-rise CFS framed composite shear wall structure was established,and a one-dimensional one-way cellular automata model was proposed to predict failure modes of the CFS framed composite shear wall structure of this kind.The inter-story shear force of the structure under the action of the horizontal seismic excitations was calculated by the mode-decomposition response spectrum method,and then the standardized inter-story shear force was defined as the cell state value of the floor cell.On this basis,the cell state similarity matching criteria was proposed,that was selecting the cells with the smallest cell state similarity coefficient as similar matching cells.The cell state similarity coefficient was calculated using Manhattan distance,and the influence of neighboring cell state values on the central cell state value in the cellular automata model was taken into consideration.According to the cell state similarity matching results,the failure mode of the known structure could be given to the predicted structure,thereby that of the predicted structure was confirmed.Finally,the CA method was used to predict the failure modes of the three types of regular CFS framed composite shear wall structures under different seismic excitations,and the results were compared with the results of time-history analysis.Some conclusions were drawn as follows: 1.For the vertical regular structure with a type of wall,the structures with the similar basic self-vibration period and lateral stiffness were selected from the known structure database to predict failure mode of this type of structure.Furthermore,the failure mode of the predicted structure was predicted by using three-types of known structure database respectively.And the prediction results showed that when the wall of the predicted structure was the same as that of the known structure,the prediction results had the smallest error,and the maximum prediction error of top floor displacement of the predicted structure was 5.43%;2.For the regular structure with two types of walls,the known structure database was built according to the different wall combinations.Then the structures having the basic self-vibration period similar to the predicted structure were selected from the known structure database.And according to the combination of the two types of walls in the floor,the floor of the known structure having the similar lateral stiffness degradation with the predicted structural floor was used for cell state similarity matching.Based on the above similarity matching results,the unknown structural failure mode was predicted,and the maximum error of the predicted displacement at the top was 2.31%.3.For a vertical irregular structure with a type of wall,the known structures which had the similar basic selfvibration period with the predicted structure were preliminarily selected.Then from the above selected known structures,the floors similar to the vertical distribution of lateral stiffness of the predicted structure were further confirmed for the cell state similarity matching.Based on cell similarity matching results,the failure mode of the predicted structure was obtained.The predicted floor which had the maximum displacement was consistent with the time history analysis result,and the maximum error of the predicted displacement at the top was 14.67%.The prediction results of failure mode of three types of CFS framed composite shear wall structures showed that the CA method was reliable and feasible for the prediction of the failure mode of regular CFS framed composite shear wall structures.