Surface settlement modelling using neural network

• Main Author: Chen, Rongxing
• Other Authors: Zhao Zhiye
• Format: Final Year Project
• Language: English
• Published: Nanyang Technological University 2020
• Subjects: Engineering::Civil engineering
• Online Access: https://hdl.handle.net/10356/145041
• Institution: Nanyang Technological University

With the development of society, the utilization rate of underground space is getting higher and higher. Tunnel projects are gradually increasing. But settlement is the most common problem in tunnel engineering. Many scholars have conducted research on this issue, and they have developed many methods for predicting the settlement. But these methods have some shortcomings, that is, they are too complicated, and the accuracy of prediction is not high. This research uses neural network to predict tunnel settlement. The neural network model can process a large amount of information like a human brain and is very suitable for non-linear problems. In this study. There are two ways of neural network analysis. The first method is to find the best performing one by building 16 different hidden neurons models. The second method is to build models with different inputs. The models with 6 input performance best while model with 4 inputs is alternate optimal model.