Example for the prediction of the unconfined compressive strength of weathered rocks using the weathering indices based on P-wave velocity

Abstract

The engineering behavior of rock materials depends not only on stress state and stress history but also on the physical and chemical change of the rock materials due to weathering. Changes in index properties such as dry density void ratio, clay content and seismic velocity caused by weathering processes is representing strength and elastic properties of rock materials. The changes caused by weathering processes in rock material may be mainly considered under the two topics; the mineralogical change and the physical change. Because the elastic wave velocity measurement is a nondestructive test, and this measurement is applied easily to repeat and be sensitive to the change of the related engineering properties. One of these parameters is Weathering State Index based on Mineralogical Change Parameter and Physical Change Parameter. Although the unconfined compressive strength of weathered rocks materials are highly important parameters for engineering geology projects, the necessary core samples cannot always be obtained from these rock materials. For this reason, the predictive models are often employed for the indirect estimation of mechanical parameters. In this study, in order to estimate of unconfined compressive strength of in rock materials, the weathering indices based on P-wave velocity and suggested early studies were used. The results of index and mechanical analysis of weathered samples used in this study were taken from the early studies for weathered magmatic rocks in the literature. And Weathering State Index was derived from the results of measurements of P-wave velocity for the said rock materials. According to the results of simple regression analyses obtained in this study, there are meaningful relationships between the unconfined compressive strength and the weathering index based on P-wave velocity. To predict the unconfined compressive strength, a fuzzy inference system having two inputs which are Weathering State Index and P wave velocity of the same samples which would have lacked pores and fissures, one outputs, the unconfined compressive strength, and 14 linguistic rules was developed. To assess the performance of the fuzzy model the following performance indexes were used namely, the variance account for VAF, and the root mean square error RMSE. As a conclusion, the fuzzy model applied in this study is suitable and meaningful model to the prediction of the unconfined compressive strength of weathered magmatic rocks using the indices. As a result, this model exhibited the most reliable predictions when compared with the simple and multiple regression models. But, it should be known that re-evaluation of the performance of the model developed in this study using additional data is useful. © SGEM2011 All Rights Reserved by the International Multidisciplinary Scientific GeoConference SGEM.