Prediction of the Uniaxial Compressive Strength of a Greywacke by Fuzzy Inference System

Abstract.

Rock engineering projects require the uniaxial compressive strength of intact rock. High quality core samples are needed for the application of uniaxial compressive strength in laboratory. In this study, to establish some predictive models taking into consideration multiple regression techniques and fuzzy inference system is aimed. Ankara greywackes is selected as the material, because of its highly problematic nature as mentioned by many previous researchers. For this purpose, a series of rock mechanics tests were carried out, and uniaxial compressive strength, point load index, block punch index, unit weight, apparent porosity, water absorption by weight, P-wave velocity, Schmidt hardness and tensile strength of greywacke were obtained. Using the obtained results, two prediction models were constructed to predict the uniaxial compressive strength of selected greywacke. The values account for and root mean square error indices were calculated as 41.49% and 15.62 the multiple regression model; 81.24% and 13.06 for the fuzzy inference system. As a result, these indices revealed that the prediction performances of the fuzzy model are higher than that of multiple regression equations.