Clustering Analysis of Gas Wells in Carbonate Reservoirs

Abstract

Capacity clustering analysis of carbonate gas reservoirs can identify groups of wells with capacity characteristics. Differentiated production management strategies can be developed for different groups to improve overall production efficiency. In order to overcome the limitations of traditional capacity categorization methods that rely on human adjustments with the inability to meet field demands in real time. This study is based on the production capacity historical data and geological data of each block in the target gas reservoir. By comparing multiple capacity clustering machine learning methods, a machine learning model for carbonate rock capacity clustering analysis based on capacity test well data was established, with a correlation coefficient of more than 0.9. The results of this study show that the predicted capacity classes can truly reflect the distribution of gas reservoir capacity. This study identifies potential high-risk gas wells for the gas reservoir, so that measures can be taken in advance to reduce the risk and improve the level of safe production.