SVNN-ANFIS approach for stability evaluation of open-pit mine slopes

Abstract

The stability of the open-pit mine slope has an important impact on the safe operation and benefits of the mining company. This paper proposes a single-valued neutrosophic number-based adaptive neuro fuzzy inference system (SVNN-ANFIS) approach to evaluate the open-pit mine slope stability. In the method, it utilizes the single-valued neutrosophic number concept for expressing the truth, indeterminate, and falsity information of influencing factors/indices (slope height, slope angle, potential slip plane angle, cohesion, internal friction angle and rock density) of the open-pit mine slope stability by truth, indeterminate, and falsity membership functions. Then, three ANFISs are used to obtain the corresponding relations between the influencing factors and the slope stability in truth, indeterminate, and falsity situations. Finally, the truth, indeterminate, and falsity output of the three ANFISs, i.e., a SVNN output, whose score value can be used to evaluate the slope stability through a score function of SVNN. The application results of the proposed approach have reflected that the training accuracy reaches 99.20% and the testing accuracy reaches 97.62%. Therefore, the established SVNN-ANFIS method provides a new effective way for evaluating open-pit mine slope stability in the uncertain environment.