The Potential Application of a New Intelligent Based Approach in Predicting the Tensile Strength of Rock
Mahdi Hasanipanah, Wengang Zhang, Danial Jahed Armaghani, Hima Nikafshan Rad
Abstract
Tensile strength (TS) of rock is one of the important properties in design process of construction civil works such as foundations and tunnels. Brazilian tensile strength (BTS) or splitting test is considered as a well-known method in evaluating TS. The present study attempts to propose a novel metaheuristic approach for the indirect measurement of BTS. This new approach is based on the firefly algorithm (FA) for training and optimizing the consequent parameters of the adaptive neuro-fuzzy inference system (ANFIS). The rock samples collected from a tunnel in Malaysia were tested in the laboratory for the purpose of providing a database consisting of totally 80 samples for analysis. Then, the statistical metrics such as coefficient of determination (R <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> ) were used to examine the acceptability of the proposed ANFIS-FA and ANFIS models. Finally, it was concluded that the ANFIS-FA (with the R <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> of 0.982) can be effectively used as a robust model to predict BTS.