Intermittent gas lift is a cyclic process which can be divided into four stages. This paper describes a mechanistic model for the intermittent gas lift system based on fundamental physical principles. The model accounts for reservoir inflow performance, fluid properties, two-phase flow in the tubing, gas flow in the tubing-casing annulus, the performance of a gas-lift valve and a surface gas injection choke, and various intermittent gas-lift control mechanisms. The mathematical model consists of a system of ordinary and partial differential equations for each of the four stages. The resulting system of equations is numerically solved. The model is capable of generating the detailed cyclic characteristics and the overall performance of an intermittent gas-lift system. The model is validated using experimental data collected by a number of previous investigators. These data cover a wide range of operating conditions for intermittent gas lift. The model's predictions compare well with the experimental measurements. The effects of various system parameters on the performance of the intermittent gas-lift system are studied using the model, and the most important controlling parameters are identified. The ability to identify these controlling parameters and their optimum values for a given system allows one to operate an intermittent gas-lift system more efficiently. Finally, the performance of intermittent and continuous gas lift are compared, and qualitative guidelines are outlined for selecting between intermittent and continuous gas-lift methods.
Intermittent gas lift has been an established oil production technique for many years. However, the unsteady-state nature of the intermittent gas-lift process has impeded developing a successful model to predict the cyclic characteristics and the behavior of the whole production system. The design of an intermittent gas-lift system is more of an art than a science and until now only "rules of thumb" have been available. Research on this topic is rare in comparison with research on continuous gas lift.
Few published works describe intermittent gas lift adequately. Most existing studies are experimental. The few predictive models developed using experimental data gathered are system dependent and cannot be applied in general. Furthermore, none of these models incorporate all the system components.
In 1962, Brown and Jessen conducted experiments on an 800 ft instrumented well equipped with 2 in. tubing. They presented a method of calculating the average bottomhole flowing pressure and the pressure stabilization time for a gas-lift cycle.
In 1963, White, et al. used dimensional analysis and similarity to model intermittent gas lift. Their mathematical simulation was simplified by assuming that the liquid slug velocity rapidly reaches a constant value and that the gas bubble penetration into the liquid slug is constant. Their experimental results compare well with the conceptual model.
In 1967, Brill, et al. reported the results of a wide range of intermittent gas-lift tests conducted in a 1500 ft experimental well. They developed an empirical fall-back correlation. The correlation was used in a conceptual model of basic fluid flow equations. Again, the model predictions compare favorably with the test data.
In 1972, Doerr conducted a study of liquid loss in an intermittent gas-lift system with 0.95 in. tubing.