Pressure Transient Analysis for Fracture-Cavity Carbonate Reservoirs With Large-Scale Fractures–Caves in Series Connection

Abstract

Abstract Fracture-cavity carbonate reservoirs are noncontinuous and highly heterogeneous with the present of large-scale fractures, cavities, and erosion pores. The multiscale and diversity of medium in reservoirs have a noticeable influence on the transient pressure analysis. In the typical curve matching, the connection between large-scale fractures and cavities must be considered. Unfortunately, the current well-testing model cannot be used to analyze the date of fracture-cavity carbonate reservoirs. The aim of this study is to develop an efficient well-testing model to obtain parameters and number of large-scale fractures and cavities. To solve the aforementioned problem, a pressure-transient analysis model for fracture-cavity carbonate reservoir with radial composite reservoir, multiscale fractures–caves in series connection, and dual-porosity medium (fracture and erosion pore) is established in this study. Laplace transformation is used to solve the mathematical model. The linear flow in the main fractures and the radial flow of caves drainage area are solved by coupling. The pressure-transient curves of the bottom hole have been obtained with the numerical inversion algorithms. The typical curves for the well-testing model are drawn. Nine flow stages and variation characteristics of typical curves are analyzed. The sensitivity analyses for different parameters are carried out. From the calculations, with the increasing in the length of large-scale fractures, the duration of linear flow is increased. While the radius of cave is the bigger, the convex and concave are the larger. Therefore, the location and the size of caves and large-scale fractures can be analyzed. As a field example, actual test data are analyzed by the established model. An efficient well-testing model is developed, and it can be used to interpret the actual pressure data for fracture-cavity carbonate reservoirs.