Highway Tunnel Surrounding Discrimination And Support The Design Of Artificial Neural Network Method

Highway tunnel is a category of underground engineering. The classification of rock masses is in the base of tunnel design and construction. It is difficult to determine the geological structure character and rock masses' character, which makes the design of tunnel support and the construction technique difficult. From the point of structural mechanics, the relation of tunnel and rock masses is based on the medium and structure's coupling, but their effect on one another is not got easily because of the infection in rock masses' geography, support style, structure style, theory mode and account parameters. This situation is difficult to fit the engineering demands and even be defeated, and at present the tunnel engineering support is based on engineering analogy.In this dissertation, thanks to artificial neural network's dealing with non-linear complicated relation, artificial neural network is used in highway tunnel's rock masses distinguish and support design, and mainly discuss the preferences, stylebook collection and disposal and how to structure improved BP neural network model. By training and check up of the stylebook, it is feasible to determine the highway tunnel's lock masses and support design.On the basis of present research consequences and the use of MATLAB language, the visual software of highway tunnel's rock masses distinguish and support design is made. The software has a good interface and can be easily used. Based on this dissertation, we get a new method to the highway tunnel design.