Evaluation of fractured carbonate reservoir and prediction of favorable areas in the eastern area of Amu Darya Right Bank

Abstract

Read online

Fractured carbonate reservoirs are significantly developed in the eastern area of the Amu Darya Right Bank. However, their types, distributions, and fracture characteristics remain unclear. This uncertainty complicates reservoir prediction and hampers exploration and development processes. Given the strong correlation between fracture development and productivity, analyzing fractures is crucial. Comprehensive evaluation and prediction methods for fractured reservoirs are essential for advancing the oil and gas industry. Based on core and geological data analyses, it finds that these reservoirs exhibit low porosity and low to ultra-low permeability. By employing conventional logging alongside specialized methods, such as electrical imaging, nuclear magnetic resonance, and far detection logging, fractures and their effectiveness can be identified and evaluated, clarifying the characteristics of reservoir spaces. Constrained by the results from core and logging analyses, seismic single attribute analysis techniques is applied to predict fractures in the HX block of Amu Darya. To mitigate the limitations of single-attribute analysis, utilize a well-supervised BP neural network method for comprehensive fracture prediction. This multi-attribute approach increases the fracture prediction probability from less than 70%–72.7%. By integrating geological understanding and well logging, and considering the influence of lithology and structure on the reservoir, synthesize the fracture prediction results to optimally select favorable areas.

Keywords

• fractured reservoir
• fracture effectiveness
• control factors
• single attribute prediction
• neural network
• favorable area selection