Tunnel boring machines (TBMs) are designed to excavate underground spaces and widely used in tunneling, civil and mining projects. TBM performance prediction substantially deals with the evaluation of machine’s penetration rate and the number of consumed disc cutters. There are various methods and equations to predict the TBMs performance in the literature. In this paper, we predicted the penetration rate and number of consumed disc cutters in Beheshtabad water conveyance tunneling project, one of the major water conveyance tunneling projects in Iran, using Artificial Neural Network (ANN) and Support Vector Machine (SVM) methods. Results showed that both approaches are very effective but SVM yields more precise and realistic findings than ANN.
Mining affects the environment around the extracted area negatively. As a result, it is vital to rehabilitate mines and reclamate the region to its former natural condition or use the land in an optimized way. The choice of herbal types and planting them to conserve the natural environment of the region and reclamating the mine is of utmost importance. When a case study was conducted in Sarcheshmeh copper mine which is one of the 10 biggest copper mines of the world, plant species were selected according to the initial factors of the rehabilitation scheme, climatic conditions and soil conditions. Then, secondary factors were identified and decision matrices that are based on questionnaires completed by experts, plant species based on regional perspective criteria, resistance to disease and insects, their strength and development, access to the plant species, economic efficiency, soil conservation and water saving, prevention of various types of contamination were all classified according to the multi-criteria decision making methods. It is worth saying that the weights of the criteria are compared with the use of PROMETHEE and Fuzzy TOPSIS methods. The best plant species in the mining area and tailing dam prioritized by PROMETHEE method are mountain almonds (2.28), tamarisk (1.44), ephedra (0.55), pistacia (0.44), Astragalus (–2.22), and salsola (–2.48).
One of the most important issues in mechanized excavating is to predict the TBM penetration rate. Understanding the factors influencing the rate of penetration is important, which allows for a more accurate estimation of the stopping and excavating times and operating costs. In this study, Input and output parameters including Uniaxial Compressive Strength (UCS), Brazilian Tensile Strength (BTS), Peak Slope Index (PSI), Distance between Planes of Weakness (DPW), Alpha angle and Rate of Penetration (ROP) (m/hr) in the Queens Water Tunnel using support vector machine .Results showed that prediction of penetration rate for Support Vector Machine (SVM) method is R2 = 0.9678 and RMSE = 0.064778, According to the results, Support Vector Machine (SVM) is effective and has high accuracy.
Purpose. Disc cutters are the main cutting tools for the Tunnel Boring Machines (TBMs). Prediction of the number of consumed disc cutters of TBMs is one of the most significant factors in the tunneling projects. Choosing the right model for predicting the number of consumed disc cutters in mechanized tunneling projects has been the most important mechanized tunneling topics in recent years. Methods. In this research, the prediction of the number of consumed disc cutters considering machine and ground conditions such as Power (KW), Revolutions per minute (RPM) (Cycle/Min), Thrust per Cutter (KN), Geological Strength Index (GSI) in the Sabzkooh water conveyance tunnel has been conducted by multiple linear regression analysis and multiple nonlinear regression, Gene Expression Programming (GEP) method and Support Vector Machine (SVM) approaches. Findings. Results showed that the number of consumed disc cutters for linear regression method is R2 = 0.95 and RMSE = 0.83, nonlinear regression method is – R2 = 0.95 and RMSE = 0.84, Gene Expression Programming (GEP) method is – R2 = 0.94 and RMSE = 0.95, Support Vector Machine (SVM) method is – R2 = 0.98 and RMSE = 0.45. Originality. During the analyses, in order to evaluate the accuracy and efficiency of predictive models, the coefficient of determination (R2) and root mean square error (RMSE) have been used. Practical implications. Results demonstrated that all four methods are effective and have high accuracy but the method of support vector machine has a special superiority over other methods.