Strength reduction strategy for rock slope stability using the variation principle based on the Hoek–Brown failure criterion

Rotational failure may exhibit in homogenous rock slopes, and the critical sliding surface of a rock slope based on the rotational failure mechanism is not a single log-spiral under nonlinear failure criterion. This work proposed an advanced nonlinear analysis method for estimating the seismic stability of homogenous slope in rock masses which is governed by the generalized Hoek–Brown failure criterion. According to the virtual power principle, the critical slope height is obtained using particle swarm algorithm. Strength reduction technique is further introduced to explore the reduction strategy of the strength parameters of the Hoek–Brown failure criterion. The outcomes indicate that the factor of safety obtained by simultaneously reducing the unconfined compressive strength σ_ci and material parameter m_i is in good agreement with the results of other methods. In addition, two cases are re-analyzed to illustrate the applicability of the proposed method, and the maximum discrepancy with existing results is about 8%. The effects of strength parameters, slope angle, and seismic quasi-static coefficients on the slope stability factor and critical sliding surface are analyzed, which shows that the seismic load and the Hoek–Brown parameters have a significant effect on the slope stability factor. There is no need to assume the expression of the potential sliding surface, which can provide theoretical support and a useful reference for the nonlinear analysis of slope stability.