Research On Elastic Property Prediction And Failure Behavior Of 2D-C/SiC Composites

C/SiC composites are widely used in aerospace field because of their high temperature resistance,high specific stiffness and oxidation resistance.However,the initial defects and oxidation damage in the manufacturing process may lead to strong randomness of the micro structure of C/SiC composites,which results in the nonlinearity of damage degradation behavior and uncertainties of mechanical properties of the material,and it greatly limits the application of the materials.Therefore,it is of important theoretical and practical significance to study the micro properties and damage behavior of C/SiC composites.This dissertation,aims at the nonlinear and uncertain mechanical behavior of 2D woven C/SiC composites,and the elastic properties and failure process of composites are studied at the micro scale and macro scale,respectively.At the micro scale,a unit cell model is established,in which the distribution of fiber bundles,matrix and holes are considered,and then the elastic properties of 2D woven C/SiC composites are studied.The unit cell voxel model at the micro scale is established by Tex Gen,in which the parameters such as fiber bundle path,section and material orientation are considered,and the model has both micro geometry characteristics of materials and meshing quality.Some elements in the matrix are randomly selected as void elements to simulate the void distribution in 2D woven C/SiC composites,and the periodic boundary conditions are imposed to form a finite element model to predict initial elastic properties of materials.Then,considering the loss of elastic properties of fiber bundles and matrix during preparation,the corresponding relations between material properties,porosity,volume fraction and initial tensile modulus of materials are established,and the combination formula of modulus prediction of 2D C/SiC composites under the influence of multi-variables is proposed by using the mixture law of composites.Considering the mutual influence of contribution value of macro modulus under the interaction of multi-variables,the modified formula of modulus prediction is given based on the combination formula,and the calculation accuracy is further improved.This method relies on a small amount of finite element calculation,with simple form and high accuracy,which can comprehensively consider the factors affecting modulus and establish the corresponding relationship between component properties,micro structure and macro modulus of the material.According to the prediction formula of modulus and test data,the damage degree of elastic properties of fiber bundles can also be given,which provides some reference for the difficult problem that the in-situ properties of fiber bundles can’t be measured at present.At the macro scale,the random distribution of mechanical parameters such as structural strength is considered,and a phenomenological constitutive model combined with the test data is established to study the failure process of structures.Through the method of polynomial fitting,it is concluded that the damage process of material tangent modulus can be divided into four stages,including crack propagation,matrix failure,fiber fracture and fiber pullout.In order to solve the problem that the analysis process at the micro scale is complex and not easy to be integrated into the macro analysis,considering that mechanical parameters of the material change with the structural position,the random distribution function is imposed to describe the random field of mechanical parameters,so as to characterize randomness of the micro structure of materials.According to the shear lag theory,a two-parameter expression is given to describe the change of materials’ residual modulus.Combined with the test data,the phenomenological constitutive equation under plane stress is established,and the Tsai-Wu failure criterion is used to simulate failure process of the structure under uniaxial tensile load.This method can be used to analyze the structure of 2D C/SiC composites at the macro scale,providing a basis for the prediction of mechanical properties and structural design optimization.