A modified three-dimensional Hoek-Brown criterion for intact rocks and jointed rock masses

Accurate description of the failure strength behaviors of rock materials, including intact rocks and jointed rock masses, is essential for engineering design and construction. First, a novel three-dimensional (3D) version of the Hoek-Brown (HB) criterion for intact rocks is proposed in this paper. A stress weighting factor n is used in this criterion to describe the effects of intermediate and minimum principal stresses. The proposed 3D version is validated using six sets of polyaxial test data, and its prediction effect is compared with that of five other existing 3D criteria. Results show that the proposed criterion exhibits the smallest prediction error for most rock types. The fitted n is closely correlated to both the partial correlation factors of intermediate and minimum principal stresses. Then, an empirical relationship m(b)(beta) between the material parameter m(b) and joint dip angle beta is developed to apply the proposed criterion to jointed rock masses. The prediction performance of the proposed empirical relation and three other existing expressions for the m(b) of six jointed rock masses at different dip angles is compared, and the proposed relation exhibits the best. The performance of the proposed criterion with empirical relation m(b)(beta) is also verified with nine sets of conventional and true triaxial test data. Results indicate that the predicted strengths are in agreement with the test data. The expression form of the established relation m(b)(beta) can also accurately describe the variation in the m(b) value with dip direction alpha.

Automated summary

Accurate description of the failure strength behaviors of rock materials, including intact rocks and jointed rock masses, is crucial for engineering design and construction. This paper proposes a novel three-dimensional (3D) version of the Hoek-Brown (HB) criterion for intact rocks, which incorporates the effects of intermediate and minimum principal stresses. The proposed criterion is validated and compared with existing criteria, showing the smallest prediction error for most rock types. Additionally, an empirical relationship between the material parameter and joint dip angle is developed for jointed rock masses, with the proposed relation exhibiting the best prediction performance.