An Improved Method to Discriminate the Carbonate Reservoir Types Combing the Empirical Model Decomposition and Energy Entropy Classification

Abstract

Carbonate reservoir is widely developed over the world, accounting for 60% of the total global oil and gas production. Due to the variation of sedimentary facies and diagenetic modifications, the carbonate reservoirs are characterized by high degree of microscopic heterogeneity. There are many reservoir pore spaces developed in carbonate reservoirs, including large karst caves, karrens, dissolved pores, fractures, intergranular dissolved pores, and intragranular dissolved pores. Conventional logging response characteristics of the various pore systems are similar. Therefore, it is difficult to identify the type of pore systems by using conventional well logs. The Cambrian - Ordovician carbonate reservoirs are the main hydrocarbon bearing layers in the Tahe oilfield, Tarim basin. However, these carbonate reservoirs are characterized by low porosity, low permeability and strong heterogeneity due to the multiple episodes of tectonic activities and diagenetic modifications they had experienced. Outcrop observations, cores, well logs and multi-scale data were used in this study to clarify the carbonate reservoir types in the Ordovician carbonates of Tahe oilfiled. Three kinds of reservoirs were divided: cavern type, fissure-cavern type, and micro-pore type. Microscopic and macroscopic characteristics of various carbonate reservoirs, and their corresponding logging responses are described in the paper. Conventional logging data is decomposed into multiple band sets of intrinsic mode functions using empirical mode decomposition method. The energy entropy of each log curves is then studied. Based on the decomposition results, the characteristics of each type of reservoir are described. Finally, by using Fisher discriminant, the types of carbonate reservoirs could be identified. Comparing with conventional logging identification methods, the method proposed in this paper avoids the instability of single-layer logging data, and increases the accuracy of logging interpretation. The method was applied to a total of 213 reservoir intervals from 146 wells in Tahe oil field. The results show that the proposed method has better identification capability than the conventional method. Prediction of reservoir types in carbonate reservoirs remain a great challenge for the hydrocarbon exploration and development. Using the methods proposed in this study, the accuracy of the reservoir type prediction improved.

AAPG Datapages/Search and Discovery Article #90291 ©2017 AAPG Annual Convention and Exhibition, Houston, Texas, April 2-5, 2017