No abstract

No abstract

No abstract

As the construction continues, the structural system and load of the cable-stayed bridge will change. At different construction stages, the forces and deformations of the structure will have their own characteristics. In this paper, in order to study the static characteristics of steel-concrete composite beam cable-stayed bridge during construction, the Jiqi Yellow River Highway Bridge is used as the engineering background, and the finite element modeling is carried out by midas. The main beam, cable tower and stay cable of the cable-stayed bridge are analyzed respectively. Among them, the static characteristics of steel and concrete bridge deck during construction are specially analyzed. In this way, it is ensured that all structures are in a safe state during the construction process, and then the construction.

Aiming at the problem of insensitivity of damage identification in conventional bridges, the application of BP neural network in long-span bridges is studied, and it is extended to the field of damage identification of conventional bridges. The finite element models of three-span continuous variable cross-section box girders under intact and damaged conditions are established by Midas civil, and the eigenvalues of bridges under different conditions are analyzed. It is found that in conventional bridges, the sensitivity to structural damage is: mode > vertical displacement > natural frequency. The parameterized natural frequencies and modes of structures are used as input of BP neural network, and the damage location and degree are used as output to train the neural network. Then, the damage location and degree are identified under different working conditions, and the results show that the recognition effect is not satisfactory. Analyzing the reason, the natural frequency and modal shape of the structure will change when the conventional bridge is damaged, but the deformation value is very small. When training the BP neural network, it is easy to appear over-fitting, which results in poor recognition effect. Therefore, it is difficult to identify structural damage by using BP neural network through acquiring the characteristics of conventional bridges. It is still necessary to study appropriate damage identification parameters and methods that can be applied to conventional bridges.