An ANN Model for Predicting Level Ultimate Bearing Capacity of PHC Pipe Pile

This paper establishes an artificial neural network (ANN) model for predicting the level ultimate bearing capacity of PHC pipe piles. The model is based on the analytical results from the gray correlation of various factors which affect the working performance of PHC pipe piles. Selecting the dominant factors as the sample inputs and the correlation weights as a factor to determine the network, we get the order of influent degree of each factor. The training analysis of the structure of artificial neural network model and the comparison between the predicted result and the results from the level static load test all prove that the model has good prediction accuracy and can be applied widely. Thus it provides a new method to analyze the ultimate bearing capacity of the PHC pipe piles in the actual projects and opens up a way to further explore the theory.