Prediction of seawall foundation settlement based on the improved variable dimension fraction and artificial neural network model

摘要

Prediction of the seawall foundation settlement it important to the engineering maintenance and disaster prevention. A new method based on the improved variable dimension fraction (IVDF) and artificial neural network (ANN) was presented on the example of the seawall located in Zhejiang Province of China. The settlement displacement analysis for a single point located on the seawall was performed. The analysis consists of three stages: idea of IVDF — ANN model analysis, IVDF-ANN modeling, and deformation forecast. The resnlt proves that IVDF-ANN model makes good use of the self-similarity of fractal theory and the self-learning ability of artificial neural network, and the method has a degree of applicability.