PERMEABILITY PREDICTION OF TIGHT CONGLOMERATES BY INTEGRATING FRACTAL CHARACTERISTICS AND SEISMIC MEME INVERSION: A CASE STUDY FROM THE TRIASSIC BAIKOUQUAN FORMATION, JUNGGAR BASIN, WESTERN CHINA

The TERRESTRIAL lacustrine fan delta conglomerate reservoirs in the Triassic Baikouquan Formation have been found as the major targets of hydrocarbon exploration in the Mahu sag, Junggar Basin, Western China. However, conglomerate petroleum reservoirs typically occupy lower permeability than sandstone reservoirs thereby determining effective permeability is critical in the reservoir characterization and modeling of conglomerate reservoirs. To investigate fractal features and better predict the permeability of the Baikouquan tight oil reservoir, integrated methods were employed, consisting of petrography, mercury intrusion capillary pressure (MICP), scanning electron microscopy (SEM), wireline, measured petrophysical properties, and seismic meme inversion. The results showed that pore types consist of microfractures, intercrystal micropores, residual intergranular pores, and dissolution pores. The fractal dimension (Fd) of the reservoir ranged from 2.15 to 2.66 with 2.46 on average. An innovative conglomerate reservoir permeability prediction model is proposed based on Fd and Kozeny–Carman equation, predicting the spatial distribution of permeability. The permeability of wells in the study was predicted with an average absolute error of 0.03 mD. Consequently, field application demonstrates that the permeability distribution combining the permeability prediction model and seismic meme inversion matches well with the measured permeability in the Baikouquan Formation. The essence of this study is that we present a method that is both scientific and effective to predict permeability with a series of data considering the fractal characteristic of the complex reservoir pore system. This promotes a deeper understanding between reservoir permeability and its fractal nature.