Statistic-chaos Hybrid Forecasting Model of Reservoir Embankment Slope under Multiple Factors Action
Chongshi Gu · Water Resources and Power · 2008
In this paper reservoir embankment slope's rainfall infiltration sliding displacement is studied based on suction deformation,relationship between water content and strain of unsaturated soil,timing effect influence factor is brought forward based on typical creep formation of elastic-viscoelastic-viscoplastic model,which offers theoretical bases for factors' choice of statistic model,and enhanced multiple factors deformation monitoring model of reservoir embankment slope is established.At the same time,aiming at some shortcomings for statistic regression model,namely fitting precision is high but extension effect is of a sort,the statistic-chaos hybridforecasting model is established for forecasting reservoir embankment's deformation tendency based on enhanced multiple factors deformation monitoring model and chaotic theory.The engineering case study indicates that the new model could enhance regression precision relative to traditional statistic model and has big practical and popularizing value for forecasting slope deformation.