| Carbon fiber reinforced Pyrolytic carbon matrix composites(C/C)have excellent properties such as low density,high specific strength,high specific modulus,high temperature resistance and thermal shock resistance,which have been indispensable materials for advanced aerospace vehicles and their power systems.Chemical vapor infiltration(CVI)process has been proved to be one of the main processes for preparing high performance C/C composites.Generally,affected by the parameters of CVI process such as preform structure,type of precursor,reaction temperature,pressure and gas transport medium etc.,C/C composites have complex microstructures.Besides fiber phase and pyrolytic carbon matrix with different textures,there are also some defects such as pores and microcracks.These microstructures are characterized as random distribution,irregular shape and free orientation.The mechanical properties of C/C composites are closely related to these complex microstructures.If there can establish the direct relationship between these complex microstructures and the mechanical properties of C/C composites,it can numerically predict the mechanical properties of C/C composites based on micro-information,which can not only provide theoretical basis for the design process of C/C composite structural,but also give parameter guidance for the design and manufacture of this material.It is expected to provide important theoretical research significance and engineering value.C/C composites can be defined as are as inhomogeneous material properties.Traditional experiments for exploring the relationship between the microstructure of composites and their mechanical properties need more expensive costing,complex testing equipment and long testing cycle.On the other hand,due to the complex structures of defect such as voids and microcracks,it is difficult for current numerical models to introduce them into the evaluation of mechanical properties of C/C composites.In this paper,on the basis of the microcosmic image of C/C composites,the geometric information of defect structures are extracted using the computer image processing method.According to the structural characteristics of composites,an appropriate mechanical model is established to realize the accurate prediction of mechanical properties of C/C composites with considering the microstructures of defect.Meanwhile,the effects of the parameters of defect on mechanical properties of C/C composites are explored in detail.The main researches are listed as:The microstructure characterization of C/C composites fabricated by CVI process.Via CVI process,C/C composites with different preform structures are fabricated.The microstruc-ture of C/C composites are observed and photographed by optical instrument,following by the volume fraction,geometric characteristics and size distribution of different components of C/C composites in the micro image are extracted and analyzed statistically on the basis of computer image processing technology.It makes a foundation for the subsequent generation of mechanical model of C/C composite.The experimental tests of the mechanical properties of C/C composite and its components materials.According to the mechanical test standard of C/C composite and its components,prepare the corresponding samples and carry out the experimental tests.Propose a method to obtain the real-time crack tip growth length during the fracture toughness test without additional digital optical aids.According to the strength distribution function of the monofilament in the bundle and experimental data,the mathematical expression of the strength distribution in the uniaxial tensile test of the fiber bundle in the C/C composite is derived.The test results can not only define the material properties of each component in the mechanical model of C/C composite,but also provide experimental verification for the subsequent numerical prediction results.The numerical prediction of elastic properties of C/C composites with considering the defects.Generate the representative volume element(RVE)of C/C composites by a modified sequential random adsorption algorithm on the basis of the of the microstructure characterization results.Propose the ”nearest distance search” method to efficiency judge the non-contact constraints of defect-defect inclusion and defect-fiber inclusion.By applying periodic boundary conditions,the RVEs are solved by finite element method,and the effective elastic properties are obtained by using numerical homogenization method.The accuracy of the strategy is verified by the experimental tests,and the effects of different defect parameters on the effective elastic properties of C/C composites is further discussed.The numerical prediction of elastic properties of C/C composites based on defect conformal mapping.Using numerical conformal mapping method to define the defect contour geometry within C/C composite micro image.Combining with the modified sequential random adsorption algorithm,generate the RVE with considering the defect shape features.Then obtain the elastic properties by solving the RVE with finite element method and numerical homogenization method.The validity of the model is verified by comparing with the experimental tests.Study the effect of the number of mapping points on the convergence of the calculation results.Considering both the accuracy of numerical calculation and the calculation efficiency of the model,discuss the geometric dimensions of RVE on the effective elastic properties of C/C composites.Meanwhile,the effects of defect volume fraction and distribution on the transverse isotropy of effective elastic properties of C/C composite are investigated.The development of cohesive element of the interlaminar fracture process of C/C composites.Based on the commercial finite element software Abaqus and Python script,develop a zero-thickness cohesive element for characterizing the fracture process zone of interlaminar cracks in C/C composites and determine the constitutive parameters and integral form of this cohesive element.Analyze the minimum critical number of zero-thickness cohesive elements required in the unit length of the fracture process zone.Solves the convergence problem of implicit solution in the calculation of delamination fracture of C/C composites by replacing the damage factor with viscosity damage by regularization method.Compared with the experimental results,it is verified that the zero-thickness cohesive elements can accurately predict the interlaminar fracture of C / C composite.Numerical evaluation of the C/C composite progressive damage considering defect microstructures.Based on the Tasi-Wu tensor polynomials failure criterion and cohesive elements,generate the stiffness degradation program of C/C composite under different damage modes.Develop the user defined subroutine USDFLD to describe the loads-material properties laws for the progressive damage of C/C composite.The numerical prediction of the effective strength of C/C composite with defects is realized,and the accuracy of the model is activated by comparing with the experimental results.In addition,the effect of defect density on ultimate strength and failure mode of C/C composite are investigated by the calculation results of different defect volume fraction. |