Application of Bayesian theory to predict permeability of reservoir

The Bayesian theory was applied to predict the distribution of hydraulic units in the Jingguan2 Block of Jinganbao oil-bearing structure belt, the permeability model was built based on the flow zone index, and the rationality of predictions was validated. The bin average, cumulative probability and exponential monotonous increase methods were applied on logs to constructure the intersection bin of 2-dimensional frequency in calculating process, and the hydraulic units of uncored wells were predicted based on the posterior probabilities of intersection bins. The best method was selected under different conditions and the permeability was predicted, and the horizontal and vertical distribution of model was validated for real reservoir conditions. The practical application shows that combining Bayesian theory with geology and well logging can not only effectively predict the hydraulic unit and permeability of reservoir, but also increase the accuracy by choosing the outstanding prediction result under different conditions, which provides more accurate geological information for reservoir description and characterization.