Grouting in rock cavern using data mining 1

• Main Author: Nur Shafiyyah Suhaimi
• Other Authors: Zhao Zhiye
• Format: Final Year Project
• Language: English
• Published: Nanyang Technological University 2021
• Subjects: Engineering::Civil engineering
• Online Access: https://hdl.handle.net/10356/150092
• Institution: Nanyang Technological University

Ground excavation, especially for underground caverns, is a more common sight these days as more underground spaces are being utilised to cope with decreasing land area. Grouting is then essential in preventing excessive water seepage, water level drawdown and maintaining the hydraulic gradient around the cavern. An exact amount of grouting should be designed in order to avoid excessive grouting and high post-grouting cost due to insufficient grout. Past studies have shown that a more analytical and accurate approach to grout design is lacking. In this study, we propose to develop predictive models for grouting volume using a data-mining process called Artificial Neural Network (ANN). Key input parameters such as flow rate, Q value, location of drilling where water seepage starts and water ingress pressure will be used to generate the predictive models and obtain a more accurate grout volume. After a thorough neural network analysis, the grout volume at each individual station was found to be closely correlated to all four parameters that are a part of hydrogeological and geological conditions. The correlation coefficient value between the input parameters and the output were also found to be significantly affected by the engineering data. Improvements are then being made to ensure that future analysis will be carried out in a more accurate manner.