The Research And Application For Reservoir Fluid Identification Based On Constituent Porous Medium Theory

In recent years, many geophysicists have been looking for new fluid identification methods for directly detecting oil and gas reservoir, in order to decrease the risk of oil and gas exploration and improve reservoir drilling rate. Fluid sensitivity factor is one of important methods of reservoir identification. Based on components fluid porous medium theory presented by professor Niu Binhua, by the application of fluid density inversion and predication technique in high porosity sandstone-mudstone strata in Baoshan and tight clastic rock area in west Sichuan, this paper uses density as fluid sensitivity factor, prove the effectiveness and universal applicability of the method in reservoir prediction. The study mainly includes three aspects: build rock physical model、determine the fluid density inversion, predict reservoir.Framework and pore fluid elastic parameters inversion is an important issue of rock physics. Fluid-saturated porous media can be viewed as two components, solid and fluid components, in accordance with the particular laws to compound a type of media. Professor Niu Binhua introduced the principle of combination of the cell bodies with the concept of the critical porosity and pore fluid composition original elastic medium model theory.Based on components fluid porous medium theory, density of entirety information can be split into framework density and fluid density. Fluid density is not only indicate ratio of kinds of fluid in pore, but also used as a criterion of distinguishing gas from water. First of all, through linear regression between porosity and density of log data, fluid density can be achieved. reservoir.The probabilistic neural network with regression, discrimination and cluster function was used to calculate the whole three-dimension fluid density data by making use of fluid density and three- dimension seismic data. Priori information was developed to train the probabilistic neural network. The seismic attributes were transformed by the trained network to identify the lithologic information of reservoirs. It turns out that the method effectively reduce inversion ambiguity.Fluid density inversion and predication technique were used for predicting reservoir in in high porosity sandstone-mudstone strata in Baoshan and tight clastic rock area in west Sichuan. Firstly, fluid density was calculated. Then, in accordance with geographic information, such as formation temperature and pressure, fluid density criterion of oil-gas identification can be formulated and used for reservoir predication. The new method offers some new instruction meaning and reference opinion for oil-gas predication well location reserves calculation. The results not only prove the theory reliable, but also prove the method effective.