In Today’s World Oil is Considered as the Most Energy-Efficient Source of fuel and Energy. Therefore The study of Reserves of Oil and Underground reservoirs is very important and has been Expanding in resent desades. In the present study the Simulation of cavities of porous environment of oil was carried out and the result of this are the same conductivity of oil and water fluids in different Geometries under the difference of underground pressure tank, for use in coded programs to predict relative permeability and Absolut fluid permeation in General, according to the result of simulation for Higher-Viscosity fluids in the type of water-loving rocks, oil is more conducive to simulation geometry conditions and water is less conductive in the same conditions. Using 3D simulation, oil conductivity is more accurate than previous ones.and relative permability were predicted Using Two methods, using neural network (with 3D simulations) without using the neural networks , which is the use of previous formulas and relationships, and the corresponding diagrams Extracted and concluded that the use of the neural network to predict the water conductivity coefficient results more accurately than when using the neural network. But for oil, the use of the neural network is not sufficiently accurate in predicying oil conductivity and needs to be corrected. Key Words Fluid Flow, Porous Media, Fluid Conductance, Pore scale