Sedimentary Microfacies Prediction Based on Multi-Point Geostatistics Under the Constraint of INPEFA Curve

Xudong Wang ^1,2 Zicheng Yang ^1,2* Xibao Liu ³ Chengyuan Yuan ^1,2

• ¹ National Key laboratory of Continental Shale Oil, Northeast Petroleum University, Daqing, China
• ² Northeast Petroleum University, Daqing, China
• ³ PetroChina Jilin Oilfield Company, Songyuan, China

Oil and gas exploration in the Songliao Basin has entered a new phase, with a particular focus on the Quantou Formation in the D Oilfield for hydrocarbon research. This investigation aims to enhance the precision of stratigraphic division and sedimentary cycle delineation, and to refine the characterization of sedimentary microfacies. To achieve this, the study employs Integrated Pattern Recognition and Fuzzy Analysis (INPEFA) and multi-point geostatistics to enhance the precision of exploration efforts. The INPEFA curve is derived from natural gamma logging data of 4215 wells, using maximum entropy spectrum attribute analysis (MESA) for precise stratification and comparison. Based on sequence stratigraphy of the INPEFA curve, sedimentary microfacies of three sand groups are simulated using stochastic multi-point geostatistics and well analysis data. The simulation results, constrained by highorder compatibility algorithms, are refined through human-computer interactive processing. The results show that sand group Ⅲ is marked by swift water encroachment, with sand bodies distributed in a strip-like pattern. Following the rapid water influx of sand group Ⅲ, sand group Ⅱ experienced a gradual lacustrine regression, leading to the formation of extensive sheet-like sand deposits. Ultimately, as sand group Ⅰ reached the maximum lacustrine flooding surface, the sand bodies exhibited the broadest distribution, with a notable intensification of fluvial activity. The method predicts sedimentary microfacies under INPEFA curve constraints, which makes sequence interface identification more intuitive and improves cycle division and correlation precision. This multi-point geostatistics-based prediction image accurately reproduces river channel distribution patterns, offering high predictability and presenting a novel approach to characterizing fine sand bodies. This method predicts sedimentary microfacies distribution in dense well pattern areas of the Songliao Basin, offering a novel solution to the challenging geological problem of accurately predicting effective sand body distribution.