Research On Computational Inverse Technique Of Discrete Element Model Parameters For Materials And Its Application In Rock Fragmentation By Cutter

With the construction of "Belt and Road Initiative" and the implementation of "Made in China 2025" strategy,the development of the construction of tunnel project is rapidly.The working efficiency and life of the tunnel boring machine(TBM)are of great significance to the actual construction.Therefore,the optimization of TBM structure has become a hot research topic.Due to TBM is in contact with hard rock masses such as granite during work,the study of rock mechanical properties and failure criterion is of great scientific significance and practical value for the development of TBM structure.Based on the discontinuity and heterogeneity of rock-like brittle materials,discrete element method(DEM)has great advantages in the study of their mechanical properties.The selection of DEM model microscopic parameters plays a key role in the calculation accuracy of the DEM model.At present,the commonly used method for microscopic parameters is called as "trial and error".However,this method is very expensive,time-consuming,and the application cost is high.Therefore,it is of great practical value to seek a fast and reasonable method to determine the parameters of DEM model.In this dissertation,the computational inverse method and experiment were used to study these issues,which provides an effective way to determination of DEM model parameters.At the same time,based on the DEM parameters determined by the computational inversion method,combined with the actual project of the rock fragmentation with cutter,the optimization for the parameters of the TBM cutter based on the DEM was carried out.The main contents of this dissertation were briefly described as follows:1.In order to establish the effective DEM model of the rock fragmentation with cutter,a computational inverse method for determination the DEM model parameters of rock-like brittle materials was developed.Firstly,based on uniaxial compression experiment,the corresponding DEM model was established and the foreword problem was built.Secondly,combined with sensitivity analysis of orthogonal test and radial basis function(RBF)approximation model,an inverse problem model for parameter identification of DEM is established,then the intergeneration projection genetic algorithm(IP-GA)is used as the solver of the inverse problem.And then the DEM parameters of rock materials were identified.Finally,based on the inverse results,the DEM computations with different shape deviation samples were carried out and the calculation results were in agreement with experimental data.These results indicated that the inverse method for DEM model parameters was effective and reliable.2.In order to provide sufficient experimental data for DEM model parameter identification for brittle materials,the experimental study on brittle materials under multiple conditions was carried out.A series of tests with uniaxial compression,Brazilian splitting and three-point bending for the homogeneous granites were performed.The mechanical properties of materials and failure modes of samples under different conditions are obtained through experiments,which provide input data and verification information for parameter identification of discrete element model.3.Combining with the different experiments of brittle materials,the inverse calculation method of material DEM parameters based on multiple experiments was developed.In order to make full use of the experimental data,the inverse calculation of the model parameters based on the stage technique,the inverse calculation of the model parameters based on two experiments and the model parameters based on three experiments were carried out.At the same time,considering that there were many different conditions and a wide range of experimental sources,multi-objective optimization was applied to the computational inverse technique.The calculation results show that the inverse calculation of the model parameters based on three experiments has better calculation accuracy.4.Based on the model parameter identification results of multi-source experiments and combined with the actuality of the rock fragmentation with cutter,the optimization of the cutter parameters of TBM was studied.Based on sensitivity analysis and approximate model of support vector machine(SVM),a multi-objective optimization model of TBM parameters was established,and the optimal parameters of cutter are obtained by particle swarm optimization(PSO).In addition,the rock fragmentation efficiency was investigated by the spacing of cutter.The results show that the optimized parameters of the cutter can effectively improve the performance and efficiency of rock breaking.