Dynamic Monitoring of Low-Yielding Gas Wells by Combining Ultrasonic Sensor and HGWO-SVR Algorithm

Abstract

As gas wells enter the middle and late stages of production, they will become low-yielding gas wells due to fluid loading and insufficient formation pressure. For many years, there has been a lack of effective dynamic monitoring methods for low-yielding gas wells, and it is difficult to determine the production of each phase in each production layer, which makes further development face great uncertainty and a lack of basis for measurement adjustment. In order to solve this problem, this paper proposes an intelligent dynamic monitoring method suitable for low-yielding gas wells, which uses an ultrasonic Doppler logging instrument and machine learning algorithm as the core to obtain the output contribution of each production layer of the gas well. The intelligent dynamic monitoring method is based on the HGWO-SVR algorithm to predict the flow of each phase. The experimental data are selected for empirical analysis, and the effectiveness and accuracy of the method are verified. The research shows that this method has good application prospects and can provide strong technical support for gas reservoir production stability and development adjustment.