Self-memorization model of dynamic system for predicting nonlinear displacement of slopes

It is of great engineering significance to accurately predict the displacement of slopes for the slope stability evaluation, slope failure forecast and catastrophe control of landslides. The self-memorization principle of dynamic system is introduced into the slope displacement prediction. By treating the time series data of monitored slope displacement as the particular solution of the nonlinear dynamic model of slopes, the nonlinear ordinary differential equation of slope deformation is deduced based on the bilateral difference principle. By using the deduced nonlinear differential equation of slope displacement as a differentiation dynamic kernel, the self-memorization model for slope displacement prediction is established based on the self-memorization theory. The model is applied to modeling and predicting the deformation time series data monitored at the slope of permanent ship lock for the Three Gorges Project and the Wolongsi slope. The case studies show that the self-memorization model is valid and feasible in predicting the displacement of slopes with good modeling and prediction accuracy and better ability of predicting displacement in more time intervals, thus a new approach for predicting deformation of slopes is proposed.