NEURAL NETWORK PREDICTION OF THE HIGH-FILL ROAD FOUNDATION SETTLEMENT OF HIGHWAY

Through use of the stronger nonlinear mapping and learning ability of the back propagation neural network, the authors develop a new artificial neural network model to predict the settlement of highway foundation. This model avoids the errors caused by artificial factors during calculation since the model is established using all in-situ observation data. The results show that the model simulation matches well with in-situ observation of the highway foundation settlement, which demonstrates its applicability in the engineering practice.