Font Size: a A A

Research Of Elastic-plastic Constitutive Model Of Rock-fill Parameter Inversion Method

Posted on:2018-11-18Degree:MasterType:Thesis
Country:ChinaCandidate:T Y WangFull Text:PDF
GTID:2322330536961335Subject:Structure engineering
Abstract/Summary:PDF Full Text Request
The constitutive model and parameters are important to analyze the stress and deformation behavior of concrete faced rock-fill dams reasonably during the stage of construction,operation and earthquake period.At present,the model parameters are usually determined by the triaxial tests and the grading scale method was used to obtain the rockfill for tests.However,the effect of scale method on the model parameters was not yet clear.Some scholars inverse analysis the rockfill parameters combining with triaxial test results and according to the monitoring data of dam deformation during the construction stage.However,most of these papers calculate the deformations of rock-fill dam during construction stage?earthquake period and after earthquake with different constitution models.The relativity of model parameters lack statistical data,and the inversed parameters can only be used to verifying the inversed problem.For example,the inversed parameters of Duncan EB model?Creep model and so on,based on the monitoring data of dam deformation during the construction stage,can only be used to simulating the dam filling construction and the process of impoundment.In recent years,with the development of the constitutive theory of rock-fill,some scholars tried to establish an elastic-plastic constitutive model which can react to the characteristics of deformation and strength of rock-fill under the static?dynamic and cycle loading.With the development and improvement of monitoring method,high rock-fill dam can basically receive detailed and reliable displacement data during construction stages.Combined with unified static and dynamic elastic-plastic constitution model,inversing analysis the model parameters of rock-fill based on the monitoring data during construction stage,and predicting the dynamic response of dam under seismic load.These have certain theoretical meaning and engineering value on the seismic design and safety evaluation of high rock-fill dam.This paper adopted generalized plasticity constitutive model considering the state of particle breakage,which initial void ratio were model input parameters.This model can describe the deformation and strength characteristics of different density of rockfill under monotonic and cyclic loading conditions,including dilatation,shearing,crushing,loop hysteresis,loop dense and loop residual deformation.The model can use a set of parameters to complete the whole process analysis of the dam construction,impounding and seismic response,and it can get the residual deformation directly.This paper inverse analysis the elastic-plastic model parameters of the rock-fill,the main work is as following:(1)First,according to the sensitive analysis to choose the inverse parameters.Then,using particle swarm optimization?artificial bee colony algorithm to randomly generate a parameter vector of choosing constitutive model,and input these parameters to the finite element model to calculate the displacements of measuring points,and calculatecorresponding fitness value,which is the deviation between the calculated displacements and the measured displacements.And continually optimize the parameters until it meets corresponding requirements.(2)Using Latin hypercube sampling to choose parameters samples for finite element analysis.Then,getting the relationship samples between the sensitive parameters and the settlements of the typical measuring points.And set the response surface between the sensitive parameters and the settlements of the typical measuring points by Artificial Neural Networks to inverse analysis the sensitive parameters.This method reduces the workload of finite element numerical analysis.(3)The proposed method was applied to determine the rock-fill parameters of Zipingpu CFRD through the monitoring settlements in the construction stage.The seismic response of Zipingpu CFRD during Wenchuan earthquake was simulated by the inverse parameters and the results were compared with the actual seismic damage survey.
Keywords/Search Tags:rock-fill dam, elastic-plastic constitution, inverse analysis of parameters, algorithm, neural network
PDF Full Text Request
Related items