6616

6167

كارشناسي ارشد

مكانيك سنگ

اصفهان: دانشگاه صنعتي اصفهان، دانشكده معدن

1390

نه، 92ص.: مصور، جدول، نمودار

ص.ع. به فارسي و انگليسي

عليرضا باغبانان،احمدرضا مختاري

حميد هاشم الحسيني

30/1/91

جواد غلام نژاد، سعيد مهدوري

1396/10/12

كتابنامه

معدن

مهندسي معدن

ID6167

به فارسي و انگليسي: قابل رويت در نسخه ديجيتالي

Prediction of Permeability in Dual Porous Media by Multivariable Regression Analysis and Artificial Neural Network Mahsa J Habibi m jhabibi@mi iut ac ir Date of Submission 2011 11 26 Department of Mining Engineering Isfahan University of Technology Isfahan 84156 83111 Iran Degree M Sc Language FarsiSupervisor Alireza Baghbanan bagh110@cc iut ac ir Ahmadreza Mokhtari ar mokhtari@cc iut ac ir AbstractOne of the most important characteristic of naturally fractured rock for simulating theflow in the hydrocarbon reservoir and nuclear waste disposal is permeability The twomajor approaches in modelling are DFN method that consider the fractures permeableand omit the rock matrix conductivity and the second method is dual porosity DP models In the latter method the two overlaid media are considered rock matrix andfractures Till now just the numerical analysis are used to study dual poroush media Innumerical analysis usually all parameters kept constant to observe the influence ofchanging the only one parameter Using the artificial nueral network and multivariateregression analysis can study the effect of all parameters simultaneously According tothe literature the simultaneous effect of the all geometrical parameters such as density themacro fracture aperture grain size and the micro aperture on permeability magnitudehave not studied In this study the 2D flow field in the fractured and permeable rockmatrix is calculated using a distinct element code and the rock matrix is simulated byvoronoi tessellation 860 models of the synthetic fracture networks were generated basedon different combinations of density and three types of correlation between fracturelength and aperture take into account and voronoi size and micro fracture aperture Inorder to propose the predition model their different statistical and fractal characteristicswere measured such as fractal dimension of intersection point fluid flow channel thearea of macro fracture the projection length of fractures in the direction of pressuregradient and the mean area of voronoi and finally the flow channel between them Theresult shows that using the principle component analysis the model is as strong as thelinear regression model and also the collinearity problem was solved The correlationbetween measured and predicted data is 80 Also the correlation coefficient betweenmeasured and predicted data of nonparametric regression using raw data is 79 and withnormal data we can see the improvement of result The correlation coefficient betweenmeasured and predicted data is 87 Compare to the both method of parametric andnonparametric regression the prediction capability of nueal network is much better Thecorrelation coefficient of neural network is equal to 96 which shows the artificial neuralnetwork is more powerful in the case of distinguish of complicated relation betweenvariables and also in prediction of target parameter Keywords Dual porous media Voronoi elements Artificial neural network Multivariable regressionanalysis Nonparametric regression

عليرضا باغبانان،احمدرضا مختاري

حميد هاشم الحسيني

جواد غلام نژاد، سعيد مهدوري