Study Of The Response Of Dynamic Characteristics Of Asphalt Pavement With Cracks

Semi rigid base asphalt pavement is widely used in highway construction because of its advantages such as comfortable driving, high bearing capacity and low noise. However, the early damage of asphalt pavement, especially the problem of the cracks of the pavement structure, has become one of the most serious diseases of semi-rigid base asphalt pavement. Vehicle load is a dynamic load instead of a static one in highway transportation, which varies with time and the road surface character, making the crack of the asphalt pavement and crack propagation problems very complex.In this paper, the fracture mechanics theory is applied to finite element method and ABAQUS is used to establish two kinds of crack propagation of the base crack and surface crack under the dynamic load. In this way, the change rule of stress intensity factor K and layer structure parameters which is in the process of the crack propagation is obtained under the plane strain model and three-dimensional entity model. Meanwhile, the results of two-dimensional calculation and the three-dimensional calculation are compared and analyzed. In addition, as to base crack and surface crack, the law of the surface deflection which varies with time is analyzed under the cyclic load and the results are compared with the one which is caused under the condition of complete structure and static load. For the two kinds of cracks, the change law of the stress intensity factor with time and the influence which is caused by the different load in the process of the crack propagation are discussed. Finally, a prediction model of stress intensity factor is established by using BP neural network. Through the research on the crack of asphalt pavement under dynamic load, the main conclusions are as follows:(1) Compared with the static load, dynamic load will reduce the surface deflection of the asphalt pavement by about 23.7%. After the crack of the asphalt pavement, pavement deflection value tends to increase significantly. Meanwhile, especially with the expansion of cracks in the structure, deflection value will increase gradually.(2) The curve shape of the stress intensity factor chang ing with time is similar to the semi sine curve, but it gets the maximum value at 0.6T because of a certain lag. When it comes to base crack, initial ones are mainly controlled by K1 and terminal ones controlled by K2, which is different from the crack under the static load. When it comes to the surface crack, they are only shown in the way of shear failure and the stress intensity factor is greater than the static value.(3) The stress intensity factor in the 3D solid model is smaller than that of the two dimensional model. In addition, by increasing the thickness of surface course, base course, subbase course and the modulus of base course, subbase course as well as decreasing the modulus of surface course, the extension of the base crack can be inhibited. For surface crack, by increasing the thickness of surface course, base course, subbase course, increasing the modulus of base course, subbase course and decreasing the modulus of surface course, the further expansion of the surface crack can be controlled.(4) In the process of crack propagation, the stress intensity factor varies with the curve of load, which shows the linear variation law. As the load increases, the stress intensity factor increases.(5) There is a high precision if we use BP neural network to establish th e prediction model of asphalt pavement crack stress intensity factor. Compared with the traditional finite element method, the application of neural network greatly improves the computational efficiency and saves time.