Macro-/Meso-scale Random Characteristics Of Hydraulic Large Aggregate Concrete And Study On Numerical Simulation Method

Concrete is divided into four grades according to aggregate size in the project.As a fundamental building material of hydraulic structures,the three-graded concrete and the four-graded concrete are also called hydraulic large aggregate concrete.In the meso scale,concrete is usually regarded as a three-phase heterogeneous composite,consisting of aggregates,mortar and the interfacial transition zones(ITZs).Therefore,the randomness is the inherent property of concrete.Randomness on the meso-scale of concrete is mainly reflected in the randomness of aggregates,mortar and ITZs,such as the geometric randomness of aggregate as well as the mechanical randomness of mortar and ITZs,etc.The random characteristics mentioned above can cause the fluctuation of macro-mechanical parameters of concrete.Moreover,the evaluation index for safety and stability of hydraulic structures is uncertain.Thus,how to consider the macro-/meso-scale random characteristics of concrete is of great significance to the scientific design,and further to the safety and stability of hydraulic structures.In this paper,a series of studies are conducted on the meso-stochastic numerical simulation method of hydraulic large aggregate concrete,the influence of meso-scale random factors,the statistical characteristics of macro-mechanical parameters and engineering application,the establishment and application of macro-scale random damage model,etc.The details are summarized as follows.1.The meso-stochastic numerical simulation method is improved from two aspects of model and algorithm,and the program is coded with FORTRAN by the author.In the aspect of model,random field theory is introduced to improve the meso-stochastic numerical model.In order to consider the spatial correlation of the material parameters,and avoid the phenomenon that the parameters of ITZ element are higher than those of adjacent mortar element,the spectral representation method(SRM)is adopted to optimize the meso-scale material parameters of concrete.Thereby,the meso-stochastic numerical model and random property model by SRM are established.In the aspect of algorithm,the crack release algorithm is introduced to simulate and mark the cracking process of concrete specimens.By simulating the crack path of specimen with inclination angle(30 degree)prefabricated crack,the feasibility of the new method is verified.By comparing the experimental result of Jinping-I hydropower project,the rationality of the method is confirmed and the mechanical parameters are calibrated.2.Using the above meso-stochastic numerical simulation method,the influence of various meso-scale random factors on macro-mechanical parameters of large aggregate concrete is analyzed in turn.Random factors include four aspects mainly: a.grading of aggregates,b.shape,size and spatial position of aggregates,c.heterogeneity of mortar and ITZs,d.parameters spatially varying of mortar and ITZs.3.The statistical characteristics of macro-mechanical parameters of concrete are studied by using statistical analysis methods(statistical distribution models,goodness-of-fit tests,etc.).Based on the results of 100 meso-numerical experiments,the study is aimed to discuss the optimal distribution model of deformation parameters and strength parameters.Moreover,the engineering significance of statistical characteristics is confirmed by the stress reliability analysis of concrete gravity dam heel.4.Based on the above meso-stochastic numerical simulation method and statistical analysis method,the establishment process of macro-scale random damage model is optimized by random filed theory(SRM and cross-correlation).Specifically,relationship between randomness on the macro-and meso-scale of concrete is established by the meso-stochastic numerical simulation method and statistical analysis method.The spatial variability and correlation of parameters in damage model are simulated by SRM.And the relationship between different parameters is considered by cross-correlation.To verify the feasibility of the new model,the model is ultilized to study the randomness of hydraulic concrete structural members.