Research on the methods of splitting and prediction point by point in tight sandstone gas reservoir productivity

Abstract

Single-point productivity evaluation and prediction are of important significance for the exploration and development in a tight sandstone gas field. The method of production splitting, multiple linear regression (MLR), and support vector machine regression (SVR) was used to establish the relationship between logging data and the gas production split point-to-point in tight sandstone gas reservoirs. In this study, the western region of the Sulige area in the Ordos Basin was the object of our research. Compared with the traditional KH splitting, the KHK splitting method was better.