Conditioning Multiple-Point Geostatistical Facies Simulation on Nonlinear Flow Data Using Pilot Points Method

Abstract

Abstract A new implementation of the pilot points method is introduced for conditioning discrete multiple-point statistical (MPS) facies simulation on dynamic production data. In MPS simulation, a conceptual model of geologic connectivity, known as a training image (TI), is used to simulate instances of facies distributions that are statistically consistent with those encoded in the TI. While conditioning MPS simulation results on static hard (e.g., core) and soft (e.g., 3D seismic) measurements is relatively straightforward, calibration against nonlinear production data is nontrivial. The pilot points method is a convenient parameterization approach in which production data are used to obtain facies values at a number of strategically selected grid blocks (pilot points). An important issue in using the pilot points method is strategic placement of the pilot points in the domain. To this end, we propose combining three sources of information to generate a score map for placing the pilot points: (i) the uncertainty in facies distribution, (ii) sensitivity information, and (iii) production data. Once the pilot points are located, the facies values at these points are inferred from production data and are used, along with available hard data at well locations, to simulate a new set of conditional facies realizations. While facies estimation at the pilot points can be performed using different inversion algorithms, in this study the ensemble smoother with multiple data assimilation (ES-MDA) is used to update permeability maps from production data, which are then used to statistically infer facies types at the pilot point locations. The developed method combines the information in the production data and the TI by using the former to infer facies values at select locations away from the wells and the latter to ensure consistent facies structure and connectivity where the production data may be inconclusive (e.g., away from measurement locations). The performance of the developed method and its important properties are evaluated and discussed using numerical experiments.