Prediction of flowback ratio and production in Sichuan shale gas reservoirs and their relationships with stimulated reservoir volume

Good flowback management in shale gas reservoirs not only preserves the conductivity of the flow paths created by the fracture network but also assists the subsequent production by minimizing the damage of the fracturing fluid to the fluid-sensitive formation. However, a quantitative prediction of its behavior and impact on the productivity of a reservoir is not available yet. This study proposed a mathematical approach to predict the flowback behavior concerning flow back ratio (FBR), and the productivity represented by first-month production (PROD), of the Changning-Weiyuan shale gas reserve in Sichuan basin, southwest China. First, a BP (Back Propagation) neural network model was established to filter out the controlling factors given the actual FBRs and the geological and engineering characteristics of seventy-six sample wells. Secondly, a correlation was generated between the FBR, the geological index g, and the engineering index e by non-linear fitting. In subsequence, the PROD was predicted based on the field recorded FBR and the comprehensive index c that combines the influences of both geological and engineering parameters. Finally, the stimulated reservoir volumes (SRVs) of twelve wells were estimated using the K-means clustering and Delaunay triangulation to assess their impact on the FBR and PROD data. It was found that the associated FBR and PROD diagrams can provide accurate predictions for the sixteen additional wells. Meanwhile, there exists an optimum FBR range of 20-40% for the shale gas region of concern where the productivity of the formation can be maximized. It is also observed that the FBR and PROD fail to display any specific trend with the value of the SRV magnitude. In general, the proposed approach can be used in shale gas reservoirs to examine the favorable FBR and guide the engineering designs to enhance production if adequate reservoir data are accessible.