Evaluating the strength and deformability properties of Misis fault breccia using artificial neural networks

Since the preparation of smooth specimens from the fault breccias are usually difficult and expensive, the development of some predictive models for the geomechanical properties of fault breccias will be useful. In this study, artificial neural networks (ANNs) analysis was applied on the data pertaining to Misis fault breccia to develop some predictive models for the uniaxial compressive strength (UCS) and elastic modulus (E) from the indirect methods. The developed ANNs models were also compared with the regression models. As a result of ANNs analysis, very good models were derived for both UCS and E estimation. It was shown that ANNs models were more reliable than the regression models. Concluding remark is that UCS and E values of Misis fault breccia can reliably be estimated from the indirect methods using ANNs analysis.