Research On The Application Of Ann In The Settlement Prediction Of Expressway Soft Roadbed

In recent years, with the rapid development of economy in China, more and more expressways have been built continuously. When the expressway is built on the soft ground, it is the key engineering problem to improve the stability and control the settlement of the roadbed. As a result, it requires higher precision of settlement prediction of soft ground embankment.In the first part of this paper, calculation and prediction methods of foundation settlement are summarized and analyzed. Then, after the brief introduction of the artificial neural network, this paper penetrates the structure, characteristics and learning algorithms of artificial neural network, and set up the BP network model to predicting the settlement process of the embankment. By comparing with the monitoring data, it proves that this method has high precision in short-term settlement prediction. Hyperbola method, three-point method and Asaoka method are used to predict the final settlement of the embankment, and the results are compared with artificial neural network.In the end, according to the monitoring data of different sections, in consideration of embankment height, thickness of crust and method of foundation treatment, a settlement correction coefficient model is set up, and then a corresponding table is made for the calculation of the final ground settlement.