Study on Facies Modeling of Tight Sandstone Reservoir Using Multi-Point Geostatistics Method Based on 2D Training Image—Case Study of Longdong Area, Ordos Basin, China

Abstract

The Longdong area in the Ordos basin is a typical fluvial reservoir with strong heterogeneity. In order to clarify the distribution law of underground reservoirs in the Longdong area, it is necessary to establish and optimize a 3D geological model to characterize the heterogeneity of reservoirs. This is of great significance for accelerating the exploitation of tight sandstone gas in the southwest of the Ordos basin. This study takes the P2h8 member of the Ct3 research area in the Longdong area as an example, analyzes the core and logging curve shape to divide the sedimentary microfacies, and establishes the facies model. In particular, in view of the difficulty in obtaining 3D training images under the existing conditions in the study area, we use the multi-point geostatistics method combining sequential two-dimensional condition simulation and the direct sampling method to establish the facies model. This method can simulate the 3D geological model by using the 2D training images composed of the digital plane facies diagrams and the well-connection facies diagrams. In addition, we choose the object-based method and sequential indicator method for comparative experiments to verify the feasibility of this method (sequential two-dimensional condition simulation combined with the direct sampling method) from many aspects. The results show that the multi-point geostatistics method based on 2D training images can not only match the well data, but also show the geometric characteristics and contact relationship of the simulation object. The distribution characteristics of sandbody thickness and modeling results are consistent with the actual geological conditions in the study area. This study explores the feasibility of this method in the 3D geological simulation of large-scale fluvial facies tight sandstone reservoirs. Additionally, it also provides a new idea and scheme for the modeling method of geologists in similar geological environments.

Employing Statistical Algorithms and Clustering Techniques to Assess Lithological Facies for Identifying Optimal Reservoir Rocks: A Case Study of the Mansouri Oilfields, SW Iran

“…Hussain, et al [21] identified reservoir rocks for lithofacies prediction by machine learning techniques. Zhang, et al [29] utilized a 2D training image of the multi-point geostatistics (MPG) method to model the facies of a tight sandstone reservoir. Liu, et al [30] analyzed influencing reservoir electrical parameters in a Quaternary mudstone reservoir containing biogas.…”

A Workflow for Interpretation of Fracture Characteristics Based on Digital Outcrop Models: A Case Study on Ebian XianFeng Profile in Sichuan Basin

“…Although fractures are extensively developed in carbonate reservoirs, fracture attributes are highly variable [2]. As effective storage spaces and permeable channels for oil/gas in carbonate reservoirs, fractures are a central topic of reservoir research; understanding fracture patterns is of great significance to oil/gas exploration and development [3][4][5][6][7][8][9][10][11][12][13][14].…”