The Triassic Baikouquan formation is one of the major strata of exploration and development in the Xinjiang Oilfield.In recent years,it has shown good potential for oil and gas resources.It is a typical low permeability sandstone conglomerate reservoir with poor physical properties and complex pore structure.The research on the pore structure and reservoir classification methods of low permeability sandstone conglomerate reservoirs is of great significance for later exploration and development in the northern part of the Mabei area.In this paper,the rock pore structure and reservoir classification methods are studied based on the core mercury injection experimental data and nuclear magnetic experimental data.The nonlinear transformation method was used to transform the NMR T2 spectrum into a capillary pressure curve,and the experimental results were used to test the transformation effect.The results show that the JSDR model method is effective.According to the shape of capillary pressure curve,the reservoirs in the study area were divided into four types,and the four pore structure parameters,including the maximum pore throat radius,average pore throat radius,porosity,and maximum mercury saturation,were used to characterize the reservoir quality and achieved good results.The four extracted pore structure parameters were taken as input.The random forest algorithm was used to classify and evaluate the reservoir.The comprehensive analysis of the reservoir classification results with logging curves and oil test data showed that this method can achieve continuous classification and evaluation of low-permeability sandy conglomerate reservoirs... |