%0 Journal Article
%T Estimating the uniaxial compressive strength of Esfandiar limestone strata based on their physical characteristics (Case study: North of Tabas City, Iran)
%J Geopersia
%I University of Tehran
%Z 2228-7817
%A Sharifi, Zohreh
%A lashkaripour, Gholamreza
%A Khanehbad, Mohammad
%A Rahimi Shahid, Mojtaba
%D 2023
%\ 09/23/2023
%V
%N
%P -
%! Estimating the uniaxial compressive strength of Esfandiar limestone strata based on their physical characteristics (Case study: North of Tabas City, Iran)
%K Physical properties
%K limestone
%K Artificial Intelligence
%K east of Iran
%K Uniaxial compressive strength
%R 10.22059/geope.2023.360469.648718
%X Uniaxial compressive strength is one of the most important properties of rocks, whose determination is important for rock engineering studies in civil engineering and mining. Determining the uniaxial compressive strength is time-consuming and expensive. In order to reduce the cost and time, you can use the empirical relationships obtained from easier methods. In this research, using the artificial neural network method, experimental relationships have been presented to estimate the uniaxial compressive strength of limestones of the Esfandiar formation in the north of Tabas city. In this method, the physical characteristics of the rock sample as independent variables are the input parameters that are used to calculate the uniaxial compressive strength as a dependent variable. These relationships consist of a general structure with 4 inputs and 1 output, which was performed using a perceptron multilayer neural network. In this research, the root mean square error (RMSE) was investigated. The results of this research show that the amount of error caused by testing, testing, and validation is close to zero and these relationships can be used to estimate the uniaxial compressive strength of limestones of the Esfandiar formation. Also, the results of the artificial neural network have been compared with the results of multivariate regression, and the results show that the value of the confidence coefficient obtained from the artificial neural network is more acceptable.
%U https://geopersia.ut.ac.ir/article_94086_a05c5ae8b21a64daa95e8f5746befb45.pdf