Slope stability analysis of the open-pit walls using artificial intelligence

Abstract

The slope stability analysis is recognized as one of the most significant issues in rock mechanics engineering. It plays a fundamental role in the design of various rock and soil structures, including mining slopes, roads, and tunnels. To date, various methods have been proposed to address the issue of stability, including limit equilibrium methods, numerical methods, and artificial intelligence techniques. In the present study, the stability analysis of mine wall slopes has been conducted using a neuro-fuzzy integrated approach (ANFIS). For this purpose, utilizing data from the Choghart iron mine, two neuro-fuzzy networks were developed to analyze the safety and stability of circular failures under static loading conditions. In the circular failure model, six parameters were identified as the most significant inputs, with the safety factor (SF) and stability (S) state as outputs, under two different scenarios for analysis. The results obtained indicate that the stability and safety analysis networks possess low error and high correlation, such that the average error for the safety factor and stability was 0.05 and 0.013, respectively, demonstrating the network's high generalization capability. Additionally, the artificial intelligence outputs of test data identified the southern wall of the mine as the most critical section, calculating the safety factor and stability of this area to be 0.81 and 0.66, respectively.

Keywords

• slope stability
• safety factor
• artificial intelligence
• fault