Abstract: Stability forecast model for high steep rocky slope was studied by Linear and nonlinear methods with the case of east open-pit high steep slope in Daye Iron Mine. Firstly, slope rockmass was classified by the composite index of rocky slope stability analysis. Based on this, forecast model for slope stability was studied by multivariate linear regression and BP neural network methods,and the results were compared with those by limit equilibrium method. The comparison results indicate that there is complex nonlinear relationship between the slope stability and its influencing factors, and it is feasible and effective to use BP neural network to forecast the stability of high steep rocky slope in open-pit mines.

Key words: Rocky slope, Stability forecast, Multivariate linear regression