Investigation of the Effect of Different Parameters on the Penetration Rate of Earth Pressure Balance Boring Machine using Fuzzy and Neuro-Fuzzy Methods, and Metaheuristic Algorithms (A Case Study: Tabriz Metro Line 2)

One of the most widely used methods for the excavation of metro tunnels is mechanized excavation using an earth pressure balance (EPB) boring machine. Predicting the penetration rate of the boring machine can significantly reduce costs in mechanized excavation. Geological and geotechnical factors, machine specifications, and operational parameters can be influential on the penetration rate of the machine. Important geotechnical factors include cohesion, friction angle, and soil shear modulus. Among the important machine parameters, the thrust force of the jacks, the torque, and the rotational speed of the cutter head can be mentioned. In this study, after analyzing the main component, eliminating the outlier data, and normalizing the data, by considering the geotechnical factors and various parameters of the mechanized boring machine, the penetration rate of the EPB machine in the Tabriz metro line 2 tunnel has been predicted. For this purpose, linear regression methods, fuzzy logic using Mamdani and Sugeno algorithms, neuro-fuzzy method, and metaheuristic algorithms were used. To validate each model, statistical indices of the coefficient of determination (), root mean squares error (RMSE), and performance indicator (VAF) were used. The results of the studies showed that the neuro-fuzzy method has a better prediction of the penetration rate in comparison to other methods. Also, the results of the sensitivity analysis revealed that the cutter head torque had the greatest effect on the penetration rate of the EPB machine.