Artificial lift

Transient Plunger-Lift Model Improves Prediction of Liquid Unloading From Gas Wells

Natural-gas wells suffer from liquid loading if the gas-flow rate is insufficient to carry liquids to the surface. Because of the technique’s many advantages, plunger lift has been used widely in gas wells for the removal of liquid columns and the rescue of dying gas wells from liquid loading.


Natural-gas wells suffer from liquid loading if the gas-flow rate is insufficient to carry liquids to the surface. Because of the technique’s many advantages, plunger lift has been used widely in gas wells for the removal of liquid columns and the rescue of dying gas wells from liquid loading. Existing plunger-lift models in the literature are imperfect either because of limited field applications or oversimplified assumptions. Several components in the cyclic movement of a plunger can be identified, with each comprising a set of specific governing equations. In this paper, a new model is described that considers the gas flow with the plunger moving in the tubing, and accounts for instant velocities during plunger rise and fall.


A typical plunger-lift system mainly comprises a piston or plunger, a surface control valve (SCV), a catcher, a lubricator or shock spring, a bumper spring, and a plunger sensor. In addition, casing and tubing serve as key components during the plunger-lifting process. Existing plunger-lift models can be categorized broadly into static and dynamic models.

Although many dynamic plunger-lift models have been studied in the literature, a universally validated model with thorough understanding of the transient behaviors of gas and liquid still is unavailable. Because of incomplete consideration of plunger dynamics or oversimplified assumptions, most existing models suffer from questionable applicability. This problem can be remedied by incorporating or proposing in-depth closure relationships and performing parametric studies to find the most-suitable one. In this study, based on previous modeling frameworks, a new transient plunger-lift model for liquid unloading in gas wells is presented. Compared with previous models, improvements have been made in the equations of plunger rising and falling velocities. The present model also accounts for different reservoir performances, which provides more-accurate and -reasonable predictions of these velocities.

Model Development

The plunger-lift model can be divided into six components: plunger upstroke, gas blowout, plunger fall, pressure buildup, gas flows above and below the plunger, and reservoir performance. Reservoir performance is a component included in all stages of the plunger-lift processes. In the energy-buildup stage, reservoir performance contributes to restoring energy below the plunger. Basic mass and momentum-­conservation equations are used to derive the plunger-lift simulation model. Model descriptions, assumptions, and related equations are discussed in detail in the complete paper.

Results and Discussions

The calculated results from the proposed plunger-lift model are presented and discussed in the complete paper, which also presents the input parameters used in the model.

Initially, the SCV is closed. The simulation begins immediately when the valve opens. The open/close time is neglected in this simulation. The calculation time is 1,000 seconds, covering a single plunger-lift cycle. The plunger accelerates as soon as the SCV opens. The acceleration of the plunger during a single cycle is not smooth; the plunger acceleration first increases and then decreases to the negative value, meaning that the plunger decelerates during upward movement. Moreover, as the slug approaches the surface, plunger acceleration increases abruptly because of the discharge of slug liquid, resulting in a spike value on a plot of plunger acceleration vs. time.

Fig. 1 presents the plunger velocity in the tubing during the complete single production cycle. The positive velocity indicates upward movement of the plunger, while the negative velocity corresponds to downward movement. Consistent with the acceleration plotting, the calculated plunger velocity first increases because of positive acceleration as the SCV opens. As the plunger moves upward, the differential pressure exerted on the plunger and the liquid slug is not sufficient to overcome the friction against the tubing wall, leading to a negative acceleration of the plunger. Thus, the plunger and the liquid slug decelerate, which reduces velocity quickly.

Fig. 1—Calculated plunger velocity in a single plunger-lift cycle.


When the liquid slug arrives at the surface, a dramatic acceleration boosts the plunger velocity quickly because of continuous depletion of liquid above the plunger, which results in a spike in the plunger velocity near 100 ft. Because the liquid slug is short compared with well depth, the acceleration is very quick, after which the plunger is caught by the lubricator and remains still. During the gas blowout, the velocity of the plunger is zero. Then, the SCV closes and the plunger drops from the top to the bottom of the well. The velocity changes to constant negative values. Finally, the plunger remains still when it drops to the bumper spring.

In considering surface production rate in a complete plunger-lift cycle, as the SCV opens, the production rate surges to a high value. However, it then drops dramatically because the blockage of the plunger and the liquid slug ceases the communication of fluids above and below the plunger, under ideal conditions. In reality, leakage exists, which needs to be considered in a strict calculation. As the plunger is caught by the lubricator, the gas blows out of the tubing and another surge in the production-rate curve can be observed. For a time, the after-flow process ends when the SCV closes. Then, the production rate decreases to zero.

In a single plunger-lift cycle, the tubing, casing, and bottomhole pressure change vs. time, contributing to the dynamics of the plunger-lift system. The production-line pressure is constant because it serves as the boundary condition to solve the model. The casing pressure changes at the same rate as the bottomhole pressure, which declines first as the SCV opens, and then increases because of the buildup after the valve closes. A dramatic drop, followed by another surge in the upward movement of the plunger, is consistent with the outflow rate at surface.

Fig. 2 shows the calculated liquid level in a single plunger-lift cycle. Initially, the liquid levels at the tubing and annulus are different because of different casing and tubing pressures. The liquid column is much longer than that seen in the annulus. As the plunger-lift cycle begins, the depletion of liquid in the annulus is very quick. However, the buildup of liquid levels in the tubing is also quicker than that seen in the annulus. As can be seen in Fig. 2, the tubing seems to always have a longer liquid column in plunger lift—meaning that liquid unloading by plunger lift is efficient.

Fig. 2—Calculated liquid level in a single plunger-lift cycle.


In this paper, a new transient plunger-lift model, based on previous work in the literature, for liquid unloading in gas wells is presented. Instead of using the constant plunger falling velocities in the tubing, the new model calculates the falling velocity in the downstroke phase on the basis of plunger gravity and drag force balance. Moreover, the model restricts the unreasonable prediction of gas-flow rate above the plunger in the upstroke phase within the gas-flow rate corresponding to in-situ sonic velocity. Future improvement of the model can be made through considering liquid leakage through the plunger and the tubing gap, liquid-slug-discharge dynamics, and multiphase flow below the plunger.

This article, written by JPT Technology Editor Chris Carpenter, contains highlights of paper IPTC 19211, “A Transient Plunger-Lift Model for Liquid Unloading From Gas Wells,” by Jianjun Zhu, SPE, Haiwen Zhu, and Qingqi Zhao, The University of Tulsa, et al., prepared for the 2019 International Petroleum Technology Conference, Beijing, 26–28 March. The paper has not been peer reviewed. Copyright 2019 International Petroleum Technology Conference. Reproduced by permission.