Research on Intelligent Optimization of Intermittent Lift Systems for Low Production Oil Wells

Abstract

Abstract In later stages of production, the depletion of formation energy can lead to a severe shortage of fluid supply, causing oil wells to operate at suboptimal production levels and reduced efficiency. Intermittent lift, as an effective solution, can enhance pumping efficiency and reduce wear and tear on both surface and downhole equipment. To establish a practical intermittent lift system, it is crucial to have a precise understanding of the dynamic fluid level and its evolution. This paper addresses the fluid level and its relationship with time-series indicator diagrams, accounting for factors like changes in inertia, friction, vibration, pressure, and fluid properties. Based on this fluid level recovery/decline law, a self-learning and self-optimizing intermittent lift system is introduced, with the goal of maintaining the liquid level within a predetermined range while maximizing daily oil production per energy consumption. The optimal design of the intermittent lift system is centered on the fluid level. Comparative analysis with existing models, such as Zhang's and Li's models, reveals a significant reduction in the average error, from 30% to 11.6%, with the implementation of this new model. Following a thorough on-site analysis, a manual intermittent lift well is selected for the implementation of intelligent design. A comparison with the manual intermittent lift system demonstrates significant improvements. The duration of switching the well on and off is reduced from days to minutes, resulting in a 4.1% increase in system efficiency, a 36% reduction in power consumption, and a 0.16t increase in oil production. This paper introduces an intelligent intermittent lift system, facilitating the transition from manual to intelligent operations while ensuring consistent fluid level adjustments. Furthermore, the system can be optimized to align with electricity pricing fluctuations, leading to cost savings and increased operational efficiency, which has practical significance in enhancing intelligent manufacturing and refining management.