Research On CFD Simulation And Prediction Of Dam Foundation Grouting

Foundation grouting is an important engineering measure to ensure the stable operation of dams and hydraulic structures.Limited by the characteristics of grouting construction,the grouting project has concealment,combined with the complexity of geological conditions and the rheology of cement grout,so it is impossible to directly observe the diffusion,penetration and filling of cement grout in the foundation,and it is impossible to directly judge the grouting construction quality.In addition,the groutability and grouting effects of rock mass are important parameters for the grouting quality.If the groutability and grouting effect can be predicted quickly,accurately and easily,it is of great significance for the control of grouting quality.Furthermore,since the regulation of the grouting pressure and the grout injection rate is complicated and often lags the occurrence of poor grouting conditions,it is difficult to ensure the quality and safety of the grouting construction process.How to solve the above problems is an important topic for the research on CFD simulation and prediction of dam foundation grouting.In response to the above problems,in-depth research has been carried out and the following innovative research results have been achieved.(1)In order to solve the problems that existing numerical studies of porous media grouting ignored the two-phase flow of Bingham cement grout and the complex geological conditions of dam bedrock,the theory and method of CFD simulation of dam foundation grouting based on 3D modeling of complex geology and VOF method are proposed,and the simulation of dam foundation grouting under complex geological conditions is realized.The existing numerical simulation of grouting involving porous media is concentrated in goaf grouting,and there is no relevant dam foundation grouting simulation research,and the characteristics of two-phase flow and the complexity of geological conditions of cement grout diffusion are not considered.Therefore,the theory and method of CFD simulation of dam foundation grouting based on 3D modeling of complex geology and VOF method are proposed.Firstly,a 3D unified model of engineering geology including different rock masses,unfavorable geological bodies and curtain grouting holes is established.Secondly,using polyhedral meshing and local mesh refinement technology,a typical computational region is selected to establish the 3D fine mesh model.The regularized parameters m of Papanastasiou equation and the numerical model are determined by the test and simulation of mini-cone slump flow and the Marsh cone.Furthermore,the cement grout is characterized by Papanastasiou regularized model,and the momentum equation of the coupled VOF method is used to simulate the curtain grouting and grout diffusion process of the dam foundation rock,and the front of the grout is also simulated.Finally,the grouting process of the bedrock grouting project of a hydropower station is simulated.The CFD technology is used to simulate the multiple-hole grouting.The curtain closure and the Bingham cement grout diffusion are analyzed,and the comparison between the simulated and the measured values of grouting time and amounts are carried out.(2)In order to solve the problems that previous studies of grouting simulation lack accurate 3D fracture network model and the appropriate CFD mathematical model in fractured rock mass,the theory and method of grouting simulation based on 3D fractured rock mass and coupled VOF are proposed to realize the fine numerical simulation of grouting in fractured rock mass.The current grouting simulation mainly focuses on simplified single fracture and 2D fracture network,or the accuracy of the constructed 3D fracture network model is insufficient,which cannot truly reflect the grout spread mechanism in 3D space;some studies of grouting simulation in fractured rock mass lack of considering the grouts as a two-phase flow,and directly embedding the Bingham constitutive equation in CFD simulations to characterize the rheological properties of cement grout will leads to difficulties in numerical simulation.Therefore,the theory and method of grouting simulation based on 3D fractured rock mass and coupled VOF method are proposed.Firstly,based on the fracture data and borehole imaging data of the exposed surface under a dam foundation,the geometric characteristic parameters of the fractures are statistically analyzed,and an improved Bootstrap(IB)sampling method is used to randomly simulate fracture parameters,this sampling method can overcome the shortcomings of the Monte-Carlo sampling method which will cause the sample collapse,then the 3D fracture network model is established using VisualGeo software.Secondly,the Papanastasiou model is used to characterize the rheological properties of cement grout for overcoming the discontinuity of the Bingham constitutive equation.A CFD grouting simulation model consisting of the Papanastasiou regularized model,VOF equation,grout surface tension and momentum resistance source is established,the numerical model is applied to the disc fracture model to verify the applicability and effectiveness.Finally,the 3D fracture network model is coupled with the CFD grouting numerical model to realize the grid representation of in the CFD calculation software.The distribution of the grout is obtained in the rock mass by solving the diffusion process of the grout in the 3D fracture network.The grouting distribution in the fracture network provides theoretical and technical basis for grouting quality control.(3)In order to solve the problem that the previous studies of groutability prediction and grouting effect prediction exist application inconvenience and insufficient prediction accuracy in fractured rock mass,an intelligent predictive model integrates hybrid grey wolf optimization(HGWO)and support vector machine(SVM)is proposed,which realizes the fast and accurate prediction of the groutability and grouting effect in fractured rock mass.The current studies of groutability prediction are mainly focused on soil grouting,while research on fracture grouting is less common,some of these studies lack comprehensive consideration of the influence factors of geological conditions,construction conditions and grout properties.Furthermore,due to the difficulty in obtaining parameters,it is difficult to apply to actual field engineering.Thus,the model and method for predicting the groutability and grouting effect of fractured rock mass based on HGWO and SVM algorithm are proposed.Firstly,a hybrid grey wolf optimization-based support vector machine(HGWO-SVM)intelligent prediction model is established,the model includes three steps: step 1,the DE algorithm and the GWO algorithm are combined to form a HGWO method with enhanced local and global search capabilities;step 2,the SVM is used as a predictive tool because it can better solve multidimensional,nonlinear and small sample size prediction problems;step 3,cross-validation and error analyses are added to the prediction model to ensure generalization performance and prediction accuracy.Secondly,based on a real-time analysis and feedback system,the input parameters such as geological conditions,construction conditions and grouting properties,and the output parameters such as the groutability are obtained,and the grouting efficiency and other parameters are obtained by the quality inspection after grouting.Finally,based on the established HGWO-SVM intelligent prediction method and the acquired data sets,judging the importance of various index factors affecting the groutability and grouting efficiency,the fast and accurate prediction of the groutability and grouting effect of the fractured rock mass can be realized.The applicability and superiority of HGWO-SVM intelligent prediction method are verified by comparison with other methods.In addition,the prediction method can reduce the field tests,improve economic efficiency,optimize the construction process,and help engineers to strengthen the quality control of grouting projects.(4)In order to solve the problems that the regulation of grouting pressure and injection rate is relatively complicated and lagging in the current grouting process,a grouting power time series prediction model based on fuzzy information granulation is proposed to provide support for the rational regulation of grouting process parameters.At present,the research on grouting prediction lacks the prediction of grouting process index,and it also fails to guide the grouting construction process in real time.Because of lacking the research of grouting power time series prediction,the regulation of grouting pressure and grout injection rate often lag poor conditions in the grouting process,thus adversely affecting the grouting project.Therefore,a grouting time series prediction model based on fuzzy information granulation is proposed.Firstly,as a control index,grouting power is introduced into the grouting real-time monitoring and analysis system,and the grouting power is collected and analyzed in real time.Secondly,a grouting time series prediction model based on fuzzy information granulation is established,consisting of three steps: step 1,for reducing the amount of data,the granular computing method is introduced to decompose the original numeric time series into a series of information granules;step 2,based on the theory of fuzzy sets,the fuzzy set operator is used to carry out fuzzy calculation for each information granular,so that the obtained fuzzy information granular can reasonably represent the original numerical point set;step 3,the support vector machine is used as a predictive tool,optimized by the grey wolf optimizer algorithm,and the information granules can be predicted quickly,accurately and robustly.Finally,based on the fuzzy information granulation time series prediction model,the grouting power fuzzy information granular,the fluctuation range and trend of grouting power during grouting process are predicted.At the same time,the prediction start time is determined.Through the performance evaluation and comparative analysis,the feasibility and effectiveness of the prediction model are verified.Therefore,the model can provide an important support for the regulation of grouting pressure and injection rate.Based on 3D geological modeling,CFD numerical simulation,heuristic algorithm,computer technology and intelligent control theory,this paper analyzes the diffusion process and grouting results of cement grout in dam foundation by CFD numerical simulation.The intelligent algorithms of hybrid grey wolf optimization algorithm,support vector machine and fuzzy information granulation are used to study the groutability prediction before grouting,the regulation of grouting pressure and grout injection rate during grouting,and the effect evaluation after grouting.The above research provides theoretical basis and technical support for dam foundation grouting.