Analysis of slope stability based on BP neural network, the analytical model of slope stability is built. Aiming at the defects that BP neural network is likely to fall into local optimum in the progress of parameter optimization; DE is associated to put forward the DE-BP neural network prediction model that is coded by real numbers. The results show that this model has a high precision for the analysis of slope stability. It is feasible and efficient to analyze slope stability based on BP neural network.

Because the crack repairing technology is varied, the crack repair application has certain complexity, and part of the project due to the repair technology unreasonable type selection. So it is necessary for all kinds of crack repairing technology was studied, on the basis of the best cost-effective choice method, crack repair engineering for reference. This study presents the basalt fiber cement mortar with BICS comprehensive repair of crack, and experimental research was done, according to the test results and the engineering experience, put forward the basic procedures and technology of crack repair method, providing reference for project application.

In the modern design and construction of building structures, building structural safety, reliability is a top priority for construction work. In combination with practical engineering quality inspection and reinforcement methods the building structure are analyzed and presented.

The vibration characteristics of the structure are important parameters of structural design if want to accurately grasp the vibration characteristics of the structure, the testing methods must be proper. At the present time, pulsation, free vibration, vibrating load are the common methods, This article adopts pulsation, knocking, initial displacement, white noise sweeping, and simple harmonic waves knocking to test the vibration characteristics of the structure and studies the characteristics of those testing methods.