Resharch On Damage And Fracture Of Concrete Subjected To Freeze-Thaw-Cycles

Most of the territory of China locate in cold region, therefore, actions due to low temperature and freeze-and-thaw cycle become the main reason of deterioration of concrete properties. It has significant theoretical and practical meaning to study the fracture damage behavior and the micro mechanism of material fracture under low temperature and freeze-and-thaw cycleIn this paper, the fracture damage problem of concrete in cold region subjected to long term freeze-and-thaw actions is studied from aspects as micro-structure, mechanism analysis, macro mechanics properties and fractal features of fracture plane after freeze-and-thaw etc. The purpose is to discover the freeze-and-thaw damage development regularities by assessing the freeze-and-thaw actions by application of fracture and damage mechanics to provide theoretical basis for the design of anti-freeze durability, assessment and service period prediction of concrete structure in cold region.Fracture tests of concrete after freeze-and-thaw were carried out to discuss the influence of freeze-and-thaw action to concrete fracture behavior and to discover deterioration features of concrete fracture parameters. The calculation formulae of concrete fracture toughness test using four point bending method, as well as calculation method of other parameters are derived by fracture mechanics principles. Under action of freeze-and-thaw, the fracture toughness, fracture energy and deterioration tendency of concrete with various water to cement ratio, air entraining agent and admixture are analyzed and discussed.There are very well linear relationship between relative dynamic elastic modulus of concrete and fracture toughness.By utilizing micro observation technology, the inner micro crack and damage features of concrete before and after freeze-and-thaw are observed. By comparison of change tendency of concrete inner micro structure, the damage mechanism of inner structure due to freeze-and-thaw action is illuminated. The optimization plan of improving concrete damage due to freeze-and-thaw is determined by comparative analysis of concrete fracture test results. The influence of various admixture on freeze-and-thaw durability is analyzed by micro observation method, and reasonable suggestion is put forward. The freeze-and-thaw damage problem is studied by damage theory and the evolution equation of concrete is established.The three-dimensional fracture surface can be precisely obtained in very short time by pasting high-strength gypsum directly on broken concrete specimen surface to facilitate high-precision segmentation and unbroken test. Fractal calculation of concrete macro and micro crack are performed by digital image process, which is better than traditional measurement method by high precision and fast calculation. This method can be applied in fractal calculation of concrete.Concrete fracture surface features very good self-resemblance, therefore, fractal theory and method can be applied to guide the test and assessment of fractal characteristics. The fractal characteristics study on concrete fracture surface before and after freeze-and-thaw provides important basis for exploring the inner regularities between fractal characteristics and fracture behavior on concrete fracture surface and establishing related mathematical models. Concrete fracture surface possesses good fractal characteristics. Fractal dimension calculation of concrete surface showes that Fractal dimension variance is very small; the relativity coefficient of linear regression is also good; and the test precision is high. So, fractal theory and fracture theory of concrete can be combined to evaluate fracture behavior of concrete under freeze-and-thaw actions.Using LS-SVM to predict fracture toughness and fracture energy of concrete under freeze-and-thaw actions doesn't need to establish equations. It is easy to build the model and modify the model. The disturbed data can be processed and it has strong summarize. Therefore, it is suitable for small sample mechine study problem. LS-SVM model obtain regularity by self learning of data files to predict tendency; however, the regularity is special and it changes with influence factors. For this reason, it is necessary to add data updating function in the model to make it possible learning new regularities that are suitable for the real condition, which keeps the high precision of the prediction model.