Evaluation of residual strength of rock based on acoustic emission data
Project Leader: Negin Yousefpour
Staff: Dr Samintha Perera
Student: Yingyan Wang, Binnan Yang, Sai Jiang
Collaborators: Dr Mehdi Pouragha Dr Jordi Baro
Primary Contact: Negin Yousefpour (email@example.com)
Keywords: artificial intelligence
Disciplines: Infrastructure Engineering
Estimating rock strength is significant in design of tunnels, mines, and other geotechnical structures supported by rock formations, as well as in other fields such as reservoir engineering, seismology and plate tectonics. This capstone project aims at predicting the post-failure (peak and residual) strength of rock formations based on their pre-failure behavior captured through Acoustic Emission (AE) sensor systems using Machine Learning methods.
AE data obtained from Uniaxial Compression strength (UCS) tests and tri-axial tests on various rock types (eg, coal, sandstone, granite) will be used to develop and validate correlations between patterns of AE energy variations and post-failure stress-strain behavior of rock samples. AE data will mainly consist of stress-strain behavior of rock samples from UCS and triaxial laboratory tests and some synthetic results obtained from Discrete Element Method (DEM) simulations.