An Embedded Modeling Method for Internal Architecture of Channelized Reservoirs

Abstract

Abstract A new internal architecture modeling method was proposed for lateral accretion(LA) shale beddings of point bar sand body in meandering channelized reservoir. Based on the point bar 3D grid model and the internal architecture analysis results, three orderly modeling technology links were applied to form a full set of embedding modeling technology process and the algorithm implementation, including 3D vector field LA pattern fitting modeling, the LA thickness distribution interpolation under LA surface trend control and partial grid subdividing model embedding process. And then, the modeling method has been validated through a typical internal architecture model of point bar at a study block of China East oilfield. The results show that firstly the internal architecture modeling method can be used to establish fine 3D model of LA shale beddings which condition to wells. Secondly, using LA pattern fitting modeling technology, the method can be effectively encountering inadequate well point information, and can predict the distribution of LA between two wells. Finally, an optimized grid description of different levels and scale architecture was provided by partial grid subdividing model embedding technology. Introduction Point bar is the most important types of sand bodies of meandering channelized reservoir. Within the point bar sand, development of LA, LA shale beddings and LA surface. Wherein, LA shale beddings are one kind of very important barrier, largely control waterflooding development, and then influence the distribution of remaining oil. Therefore, the establishment of three-dimensional quantitative internal architecture of point bar sand body model has the practical value. It can be seen from the internal architecture model of point bar, LA, LA shale beddings and LA surfaces of the three elements. Among them, LA surface as the LA of a flushing body exists between the erosion surface; the LA shale bedding refers to the LA shale layer deposited on the surface, a more oblique insertion mud wedge in the section, the plane is arc, thickness of between 0.2m~2m. Point bar LA models generally can be summarized into three kinds, respectively is the horizontal echelon (see Fig.1A), ladder echelon (see Fig.1B), wave (see Fig.1C). This thesis research algorithm will focus on horizontal echelon and step diagonal LA pattern lateral layers as modeling objects. The goal of modeling for internal architecture of point bar, is the LA shale beddings model embedded grid on every single point bar model(see Fig.2). To this end, first modeling method of embedded architecture surface model by an automatic model fitting surface modeling techniques used to characterize the spatial distribution of the LA; secondly, according to the LA laminate thickness data extraction for each well point, and in the LA surfaces within the scope of interpolation, to establish a model of LA laminate thickness; finally, on the basis of LA model surface and LA laminate thickness, and to the local grid refinement approach will be embedded into the point bar LA shale beddings solid model established in the previously step.