Slope Stability Evaluation Methods Based On Evidence Theory

Effective analysis and evaluation of slope stability can reduce and avoid some losses caused by slope landslides to a large extent.The parameter information about slope stability has many uncertainties such as randomness,ambiguity,and incompleteness,which caused by the complexities of slope structure itself,the limitations of the sensors or the observers,and the imperfection of technologies of accessing information.However,traditional slope stability analysis methods usually do not take into account the uncertainty of parameters,which leads to a certain gap between the slope stability assessment results and the facts.Aiming at the problem of parameter uncertainties,the methods of slope stability evaluation based on evidence theory are studied in this paper.The main work is as follows:(1)System reliability assessment method based on evidence reduction and random set mapping.In view of the problem of excessive calculation in the reliability evaluation and random set mapping method,the random set form(evidence)of the input parameters in the evaluation model is reduced.The reduced result is mapped to the output based on the expansion criterion to obtain the output random set form(evidence).Then,reliability evaluation can be realized by converting the output evidence into a Pignistic cumulative probability distribution.Finally,the effectiveness of the proposed method is verified in the experiment of typical nonlinear evaluation functions.(2)Slope stability evaluation method based on evidence reduction and fuzzy parameter random set mapping.Besides the random uncertainty of parameters proposed in(1),the fuzziness of slope parameters is further considered in(2).The fuzzy parameters are transformed into evidence in the form of random sets.Use Dempster combination rule to fuse and reduce fuzzy evidence proposed by different experts.These evidences are mapped to the output using the extension principles.The Pignistic cumulative probability distribution can be used to evaluate the slope stability.Finally,the effectiveness of the proposed method is verified through typical rock slope stability evaluation experiments.(3)Slope sliding force prediction and stability evaluation via belief rule-based inferential methodology.For sliding force variables that can intuitively reflect the stability of the slope,a belief rule-based(BRB)model is established to describe the non-linear and uncertain relationship between the history/current sliding force and the future sliding force.Then,evidence reasoning(ER)algorithm is adopted to fuse the activated belief rules.And based on the fused results,the sliding force at a future time can be predicted accurately.Moreover,considering the variation of the sliding force on different slops or different monitoring points in the same slope,a parameter transfer strategy of BRB model together with a corresponding online update method are proposed to achieve the adaptive design of the BRB prediction model.Finally,the effectiveness of the proposed sliding force prediction methods has been verified by the experiments on the sub-section of the China West-East Gas Pipeline Project.