Abstract:The distribution and oil content of deep carbonate oil reservoirs are extremely heterogeneous. The prediction of favorable areas for hydrocarbon accumulation in such oil reservoirs has always been the focus and difficulty of research. To address this problem, this study selected the carbonate oil reservoir developed in the Middle-Upper Ordovician in the Tuofutai area of the Tahe Oilfield in the Tarim Basin as a typical example. The formation water chemical characteristic data and single-well production data in the study area were statistically analyzed to construct a data set. Combined with the random forest algorithm and using the idea of hierarchical classification learning, a prediction model for favorable hydrocarbon accumulation areas at three levels was trained. Finally, quantitative prediction of favorable areas was conducted. The results show that the F-measure indexes of the three prediction models were 0.7310, 0.7589, and 0.6498, respectively. The most favorable areas for hydrocarbon accumulation predicted by them are in the intersection area of northeast-trending and northwest-trending faults, and the favorable and sub-favorable areas are near the single-direction fracture. The prediction results are consistent with the actual production, which proves that the prediction method has high credibility and feasibility. This study can provide a method for deep carbonate oil and gas exploration in the study area and areas with similar hydrocarbon accumulation models, and provide ideas for quantitatively predicting the favorable hydrocarbon accumulation areas of other types of oil and gas reservoirs by applying formation water chemical characteristics. It has significance for the application of machine learning methods in the exploration field.

Optimization of Seismic Attributes and Reservoir Prediction Based on Cluster Analysis——By Taking Ao 9 Working Area in Aobaota Oilfield for Example

Application Of Multi-Attribute And Neural Network To Predict Reservoir

Application of seismic multi-attribute optimization technology to the prediction of carbonate fractured-vuggy reservoir

Seismic attribute optimization method and its application for fractured-vuggy carbonate reservoir

Application of Attribute Optimization to Output Prediction of Tai'er Member Gas

Multiattribute Reservoir Prediction of Hei 49 Block in Daqingzi Oilfield, Ne China

The regional multiple seismic attribute quantitative reservoir prediction method based on sedimentary characteristics

Method of Oil/Gas Prediction Based on Optimization of Seismic Attributes and Support Vector Machine

Applications of Seismic Attribute Optimization and Inversion Prediction Technology in CJZ Eastern Slope of Shengli Oil Field

Finely Description for Fractured Reservoir and Comprehensive Evaluation of Seismic

Application of the Combination of Well and Earthquake in Reservoir Prediction of AoNan Area

Application of high-precision reservoirs prediction techniques in Zhouqingzhuang Oilfield

Application of fine seismic-geologic prediction technique in exploration of lithologic oil/gas reservoir: Case of Zhangdong area in Dagang beach

Multi-attribute optimization analysis for sandstone porosity prediction

Reservoir Prediction of Lishui Depression, East China Sea Basin

Comprehensive prediction method of seismic to thin sandstone reservoir in delta-frontal

Application of multiple seismic attributes in the prediction of hydrocarbon potential for reservoirs in Tahe oilfield.

Seismic Integrative Prediction of Fracture-Cavity Carbonate Reservoir:Taking ZG21 Well Area in Tarim Basin As an Example

A prediction method for favorable hydrocarbon accumulation areas in carbonate oil reservoirs based on machine learning and driven by formation water chemical characteristic data

Seismic Attribute Gradient Analysis and Reservoir Configuration Study of Shallow Water Delta Reservoir in Huanghekou Sag

Seismic attributes analysis and lithological reservoir exploration.