Abstract

The reservoir geological model can provide detailed reservoir characterizations for oil and gas development, thereby addressing the problems of sparse well patterns and lack of data for deep-underground reservoirs in the Yubei area. In this study, stochastic simulations based on geostatistical analyses with seismic attribute constraints were used to predict the physical properties of the Wutonggou Formation reservoirs and establish a fine 3D geological model. Log interpretation models were established to represent the longitudinal distribution of petrophysical parameters. Next, the structural framework of the strata was constructed based on seismic interpretations, and seismic attribute analysis was subsequently performed. Genetic inversion and neural network algorithms were used to obtain acoustic impedance data and predict petrophysical parameters. The planar distribution characteristics of the parameters were then clarified. Finally, petrophysical models were established using Gaussian random function simulation under the control of the facies model, which in turn was simulated via sequential indicator simulation. The model results were evaluated based on the well and production data, and the model parameters exhibited distribution patterns similar to the well data. The model parameters also corresponded well with production data, exhibiting a compliance rate of 90.98%. The model shows that the study area has a large range of reservoir depths and is generally oriented NW-SE. The Wutonggou Formation mainly developed oil-bearing reservoirs of different depositional types at the bottom of the third and first members, and the thickness of the depositions increased significantly with depth. The reservoirs, as a whole, are characterized by medium-low porosity and low permeability. Furthermore, compaction has led to significant reductions in porosity and the formation of microcracks in the glutenite of deep-buried reservoirs, which play important roles in improving reservoir permeability and hydrocarbon flow properties.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call