| The selection of foundation for building is a complex and comprehensiveness work.It needs various knowledge and professional experience for the structure design. Especially in saline soil area, bad soil selection based on the nature of increased uncertainty. In this case, deterministic method can not be achieved, but the artificial neural networks are non-deterministic method can play a role. The artificial neural network approach to salt selection of land for the bridge foundation are used in this paper, the qualities of the ANN, high-nonlinear, high-permissibility of error and high-robustness, self-adaptability, online work, and so on, are adequately used.This paper through the large foundation engineering research achievements of retrieval, Summarized the engineering characteristics of the saline soil area and engineering disease, And outlined the basic principles of BP neural network. When the model of artificial neural network about the selection of saline soil area bridge foundation is build, the advantage and disadvantage of foundation are analyzed,and the main affecting elements that can dominate selection of brideg foundations are picked out.. Then, the question on how to choose the numbers of the hidden-layer and neuron is discussed, the structure of hidden layer about BP artificial neural network is identified. The identification model of neural network about the seletion of saline soil area bridge foundation is established based on neural network toolbox of Matlab7.1 is designed. The model can help designers to easily select foundation for the saline soil area bridge.Inputting the geology date of typical project items, train network, optimizing the network model. By testing, the trained model can help designers to selece an approptiate kind of foundations. The demonstration shows that this method has an important practical significance and a referntial worth in the design of the saline soil area bridge foundation.
|
PDF Full Text Request