In this research, establishment of a good relationship between N and UCS of a rock mass under particular geological circumstances is considered. So, collected data related to the immediate roof rock of coal seams in North-Eastern coal fields of Iran are used in this paper. In order to determine the N and UCS, a significant number of samples were selected and tested, both in-situ and in the laboratory, and data of N and UCS have been collected. Then, two kinds of reliable relationships between N and UCS data are considered. Firstly, as a result of regression analysis, an equation with the best fit relationship is reviewed. Secondly, an artificial neural network (ANN) is developed using actual data sets, and showed higher accuracy. The resulted network can be used to estimate UCS of the roof rock in coal extracting areas in the mentioned zone by performing simple in-situ Schmidt hammer tests.
The UCS of rocks is one of the important input parameters used in rock engineering projects such as design of underground spaces, rock blasting, drilling, slope stability analysis, excavations and many other civil and mining operations. Testing of this mechanical property in the laboratory is a simple procedure in theory but in practice, it is among the most expensive and time-consuming tests. This calls for transportation of the rock to the laboratory, sample preparation and testing based on the international standards. In order to carry out these standard tests, special samples, such as cylindrical core or cubical samples, need to be prepared. Preparing core samples is difficult, expensive, and time-consuming. Moreover, the preparation of regular-shaped samples from weak or fractured rock masses is also difficult. Under these circumstances, the application of other simple and low-cost methods to carry out the above tasks with acceptable reliability and accuracy will be important. Therefore, indirect tests are often applicable to estimate the UCS, such as Schmidt hammer, point load index and sound velocity. Indirect tests are simple, require less preparation, and can be adapted easier than field tests.