1School of Mining Engineering, College of Engineering, University of Tehran, Tehran, Iran
2Simulation and Data Processing Laboratory, School of Mining Engineering, University of Tehran, Tehran, Iran
Water inflow is one of the most important challenges in the underground excavations. In addition to inducing working conditions and environmental problems, it decreases the stability and quality of the surrounding rocks. The direct method of measuring rock mass hydraulic conductivity consists of drilling the boreholes and observing the rate of fluid lost in the boreholes. Applying this method is still problematic due to the depth of underground spaces, and also the groundwater level covering them. Therefore, many researchers have tried to predict the water inflow indirectly. This paper attempts to predict the groundwater conditions in the Beheshtabad tunnel (in Iran) using the fuzzy inference system based on the datasets acquired from the preliminary exploration studies. 250 datasets for the Beheshtabad tunnel were used out of which, 200 datasets were used to develop the model and 50 were used to validate the results obtained. 90% accuracy was obtained through comparing the fuzzy estimation and actual groundwater conditions. The proposed model can be used with much less degree of complexity for prediction of the groundwater conditions as well as decreasing the overall costs of the exploration measurements, and due to these characteristics, it is applicable for most users.