Prediction of organic facies of deep source rocks in southwestern part of Bozhong Sag, Bohai Bay Basin

Abstract

Considering the characteristics of abundant source rock formations and heterogeneity in the southwestern part of the Bozhong Sag, Bohai Bay Basin, based on the geochemical and petrographic analysis of source rocks, the methods of ΔlogR series, multiple regression and BP neural network are applied to predict the organic matter abundance under the constraints of sedimentary facies.The BP neural network method is preferably used to predict the total organic carbon content and cracked hydrocarbon content of source rocks in a single well, and then the hydrogen index prediction curve is calculated.Combined with total organic carbon content, the organic phase of each source rock layer is quantitatively described, and the advantages and disadvantages and distribution characteristics of source rocks are analyzed. Results show that the prediction accuracy of BP neural network method is relatively higher. The source rocks in the lower section of the second member of Dongying Formation are featured by kerogen of type Ⅱ2, hydrocarbon index of 125-400 mg/g, TOC content basically less than 3%, and organic facies of BC, C and CD types. The source rocks in the third member of Dongying Formation are featured by kerogen of typesⅡ1 to Ⅱ2. The source rocks in the first, second and third members of Shahejie Formation are featured by kerogen of types Ⅰ to Ⅱ1, TOC content about 3% or higher than 3%, and organic facies of B, BC and C types.The high-quality source rocks of the Shahejie Formation are mainly developed in the central and southern parts of the study area, and are the key areas for further exploration.

Keywords

• total organic carbon content
• δlogr method
• organic facies
• source rock
• bp neural network
• bozhong sag
• bohai bay basin