摘要：The shallow-burial Neo-Tethys Cenozoic reservoirs host significant hydrocarbon resources. However, a limited understanding of the Paleogene Apollonia Formation, north Western Desert, Egypt, hinders further exploration and development. Here, RZK, WD33 and JD blocks were analyzed using a combination of petrological observations, petrophysical analysis, well logging analysis, machine learning, and deep learning. Five types of limestone and two types of dolostone were identified. Planktonic and benthic foraminifera (e.g., Nummulite) are well-developed. The sedimentary facies consists of grain shoals, shallow to deep ramp and deep-water slope facies. Grainstone and packstone from high-energy sedimentary environments yield high primary porosity and permeability. Dolomitization and bacterial sulfate reduction further enhance reservoir properties. Framboidal and granular pyrites are well-developed here. The reservoir spaces consist of framework pores, inter-granular pores, intra-granular pores, moldic pores, inter-crystalline pores, and fractures. The sedimentary facies and diagenetic history indicate that WD33 and RZK (esp. ERZK ＆ MRZK) blocks exhibit promising reservoir performance. Drilled cores of three wells from different sedimentary facies served as the training dataset for machine learning and deep learning, while two core wells were utilized as the test group. Using XGBOOST, Random Forest, and Long Short-Term Memory algorithm, the predicted porosity values closely match core-measured values, with R2 (coefficient of determination) values of 0.72, 0.58, and 0.61, respectively. The permeability prediction results also yield similarity to the measured values. The Jurassic and Cretaceous source rocks generated hydrocarbon for the overlying Paleogene Nummulite carbonate reservoirs in the Neo-Tethys domain. This study provides valuable insights into carbonate reservoirs in the north Western Desert and holds significance for global Cenozoic carbonate exploration and development.

关键字：Neo-tethys domain; Carbonate reservoir; Nummulite; Machine learning; Deep learning

ISSN号：1464-343X

卷、期、页：卷: 227

发表日期：2025-07-01

期刊分区（SCI为中科院分区）：三区

收录情况：SCI（科学引文索引印刷版）,SCIE（科学引文索引网络版）

发表期刊名称：JOURNAL OF AFRICAN EARTH SCIENCES

参与作者：Hu, Yongjie,Zhang, Hong,Liu, Zixuan,Cross, Edgar,Li, Ying,Shi, Shuyuan,Liu, Hongxia,Ramy, Emadeldin,Ma, Rongtu

第一作者：蔡春芳

论文类型：期刊论文

论文概要：Hu, Yongjie,Zhang, Hong,Liu, Zixuan,Cross, Edgar,Li, Ying,Shi, Shuyuan,Liu, Hongxia,Ramy, Emadeldin,Ma, Rongtu,蔡春芳，Reservoir evolution and property prediction of the Paleogene Apollonia Formation, Western Desert, Egypt，JOURNAL OF AFRICAN EARTH SCIENCES，2025，卷: 227

论文题目：Reservoir evolution and property prediction of the Paleogene Apollonia Formation, Western Desert, Egypt