Using the Multiple Linear Regression Method for CO2 Flooding Evaluation in the Daqing Oilfield

Abstract

CO2 flooding and burial efficiency can be improved by establishing a standard for screening suitable CO2 flooding reservoirs for the Daqing Oilfield. Moreover, the influencing factors of CO2 flooding can be classified into geological factors, fluid properties, and development factors. An evaluation index system and hierarchical structure are created based on the importance of multiple factors. The subjective analysis error of human beings is quite large when establishing the evaluation index system, especially in the fitting curves that are drawn by different analysts. Based on the geological characteristics of block Bei14 in the Daqing Oilfield, a typical CMG model is presented in this article. A total of 15 factors in the 72 models are used as independent variables, and the recovery factor is used as a dependent variable for multiple linear regression calculations. In addition to sensitivity tests based on how much significance is indicated by the t value in the results, a unique result can be calculated using standard statistical methods when analyzing the calculation results of the multiple linear regression model. The results of the screening standard evaluation system are consistent with the production history of the oilfield based on the mathematical understanding of multiple factors of CO2 flooding. Around the high-score well group, oil saturation decreases significantly, and the cumulative production is generally higher than that of the low-score well group. The calculation results of block Bei 14 show that 74% of well groups have an evaluation value greater than 0.50, and 72% of well groups have an annual oil exchange ratio above 40%, which means that over 70% of well groups can benefit from CO2 flooding. Thus, CO2 flooding can be applied in the Daqing Oilfield, and multiple linear regression can provide effective guidance for the Daqing Oilfield’s development.