The Taguchi approach to the evaluation of the influence of different testing conditions on the mechanical properties of rock

摘要

Laboratory testing is often used to derive the mechanical properties of rock. Testing conditions heavily influence the results of such laboratory experiments in which factors, including the water content, diameter of samples, slenderness of sample and strain (or loading) rates are of great importance. This paper evaluates the influences of four major test conditions: water content, strain rate, sample diameter and sample slenderness, on the peak uniaxial compressive strength (UCS) and modulus of elasticity (MoE) of sandstone. Following the Taguchi approach, an experimental study was conducted on cylindrical sandstone specimens, and the results were interpreted using signal-to-noise ratio (S/N) and analysis of variance (ANOVA). The results reveal that water content is the most influential test condition for peak UCS and the influence of sample diameter, slenderness and strain rate decreases in the cited order. MoE is greatly affected by sample slenderness, whereas the other three test conditions show an approximately similar and smaller influence. These characteristics were further verified by the ANOVA results. These behaviours are consistent with the results reported in the literature. Finally, the Taguchi approach, which is a very useful and versatile technique, which has not been effectively applied in rock mechanics and rock engineering, was successfully used to evaluate the influences of different test conditions on the peak UCS and MoE of laboratory rock samples.