Model Helps Guide Choke-Management Strategies Under Constraints

Fig. 1—Logic diagram of the choke-selection algorithm.

You have access to this full article to experience the outstanding content available to SPE members and JPT subscribers.

To ensure continued access to JPT's content, please Sign In, JOIN SPE, or Subscribe to JPT

Choke-management strategies vary significantly among operators. No rigorous methodology exists for properly selecting choke sizes when constraints are placed on completions, wellbores, and fluid pressures and velocities. In this paper, the authors propose a coupled wellbore/reservoir model that performs dynamic nodal analysis using integrated models for surface-facilities, wellbore, and reservoir simulators and allows an operator to select choke sizes as a function of time.


The suggested algorithm consists of three primary entities: the reservoir, the wellbore, and the completion. A reservoir simulator provides the rate as a function of drawdown or bottomhole pressure (BHP), and a wellbore model is used to calculate the frictional pressure loss along the wellbore and the surface flowlines. Depending on the fluid system, the appropriate choke-flow model is used. Dynamic (time-dependent) nodal analysis ensures the continuity of pressure and rates between the wellbore and reservoir entities at every timestep. The algorithm suggests the maximum available choke that satisfies, at all times, the entire set of user-­specified constraints.

The complete paper applies the method to two scenarios. The first is a high-permeability, vertical casedhole well in which screen erosion is a serious concern. For that particular case, the choke sizes should be adjusted to limit perforation velocities under the critical threshold of 10 ft/sec, which has been shown to cause screen erosion in high-permeability offshore wells. The authors show that choke sizing depends on various parameters, including the separator pressure, the well trajectory, and the true vertical depth. Finally, the method is applied to the design of a flowback operation in order to prevent proppant flowback and proppant crushing in a hydraulically fractured unconventional well.

Design Considerations

To mitigate the risk of productivity impairment or failures associated with the completion or other equipment, production engineers should take into consideration existing guidelines for allowable values of flow velocities or drawdown limits. These recommendations can be classified into three major categories: wellbore, completion, and reservoir constraints. Wellbore constraints include, but are not limited to, the maximum pressure drop across the choke to prevent hydrate or wax and asphaltene formation downstream of the choke, the maximum fluid velocity in the surface flowlines to prevent erosion, and the minimum fluid velocity along the wellbore trajectory to ensure effective proppant transport during flowback operations. Completion and reservoir constraints depend on the completion type in place. Table 1 of the complete paper presents several completion and reservoir constraints along with their maximum allowable values reported in the literature.

Model Description

To properly design a flowback operation or a choke-management strategy, reservoir, wellbore, completion, and choke-flow models must be combined. The method presented comprises two major entities: the reservoir and the wellbore. Both entities are modeled separately, and this modularity allows any commercially available reservoir simulation or wellbore model to be deployed by the algorithm presented herein.

Reservoir Model. The reservoir model contains all the properties typically used in a reservoir simulator (i.e., reservoir geometry, formation properties, and initial conditions) along with an appropriate grid capable of accurately delineating the near-wellbore region and the fracture geometry. For a given set of initial conditions (i.e., pressure and saturation distribution) and flowing BHP, the reservoir model provides the production rates and the final distribution of pressure and fluid saturations. It is important to note that the reservoir entity accounts for the reservoir only, excluding any completion model. Consequently, the BHP used as input to the reservoir model is the pore pressure at the completion/reservoir interface.

Wellbore Model. For a given choke size, the wellbore model provides the pore pressure at the completion/reservoir interface as a function of the liquid rates. The wellbore model comprises the surface equipment, the selected choke size, the wellbore trajectory, and a completion model for calculating the pressure drop along the perforations and the annular pack. Depending on the fluid system under consideration, the appropriate choke-flow model should be used.

Dynamic Nodal Analysis. To match the pore pressure at the reservoir/completion interface, a dynamic nodal-analysis scheme was used. To reduce computational effort and minimize the number of (reservoir) simulations required, a ­secant-like method was selected in lieu of the otherwise faster Newton-Raphson Method. Using the secant method, convergence is typically achieved in four to five iterations.

Choke-Selection Algorithm. The primary objective of the algorithm is to select, at all times, the largest choke size that satisfies the entire set of constraints placed on the system. In other words, the algorithm maximizes production while ensuring that wellbore, completion, and reservoir constraints are met. Fig. 1 above presents the logic diagram of the choke-selection algorithm.

The assessment of the failure criteria requires that all the necessary calculations and checks must be made until all constraints are met. When all constraints are met, the algorithm proceeds with testing the next-larger-available choke size. On the other hand, if one or more constraints are not satisfied, the currently tested choke size is considered unsuitable and the algorithm reverts to the previous smaller choke size that satisfied all constraints.

After a choke size has been selected, the algorithm will simulate the reservoir domain for a user-specified time and update the reservoir conditions and the BHP through nodal analysis. The process will terminate once the simulation is exceeded or the maximum choke has been implemented.

Model Application

The suggested choke-selection algorithm was applied to a conventional vertical well and a hydraulically fractured horizontal well. In both cases, reservoir simulations were performed with a commercial black-oil reservoir simulator. An analytical multi­phase flow model was used to simulate fluid flow along the wellbore.

Vertical Casedhole Well. In this example, a choke-management strategy is sought that satisfies a set of constraints for a given formation and production system. It is not known a priori which of the three constraints (pressure drop along perforations, perforation velocity, and annular velocity) will be crucial in the selection of choke size as a function of time.

The choke-selection algorithm was run for this case. At time=0, instead of using the smallest compatible diameter, the algorithm selects the largest choke diameter that satisfies all three constraints placed on the system. For this particular case, perforation velocity is the crucial factor that determines the choke size as a function of time.

For the same set of constraints, a smaller separator pressure requires a smaller choke diameter: The pressure difference should now be provided as friction loss across the choke. This simple, yet important, observation proves that the selection of the choke sizing depends on various components of the system such as the separator pressure.

Hydraulically Fractured Horizontal Well. In this example, a choke sequence is sought that maximizes production and mitigates the risk of excessive proppant flowback. To quantify proppant flowback, a numerical study that uses grain-scale discrete-element modeling simulations is used to assess the combined effect of effective closure stress, pore-pressure gradient, and particle size on the amount of proppant being produced from a single planar fracture.

At the beginning of the cleanup operation, the confining stress is low and the proppant can tolerate a small hydraulic pressure gradient. As the effective stress increases, a larger pressure gradient is required to destabilize the proppant pack. For simplicity, it may be assumed that the maximum (allowable) pressure gradient is a logarithmic function of the effective stress acting on the proppant.

The design of the cleanup operation requires that no more than 30% of the proppant flows back into the wellbore. For a given choke size, if the actual pressure gradient along the fracture is larger than the maximum allowable pressure gradient, then the choke is considered too large and a smaller-diameter choke should be used.

The method is applied to an unconventional oil well in the Wolfcamp B formation. Refined mesh was used in the well vicinity to obtain an accurate estimate of the pressure distribution close to the wellbore and along the fracture. To reduce computational effort, a quarter of a single planar fracture was simulated using symmetric element modeling.

The algorithm will check whether the choke size can be increased every 8 hours. This time schedule depends on the availability of personnel. The algorithm selects the choke size so that the actual pressure gradient along the fracture does not exceed the allowable pressure gradient, assuming that no more than 30% of the proppant flows back. The model captures the decline of water/oil ratio (WOR) as a function of time as well as the increase of the total liquid rate when a choke of a larger diameter is applied. The algorithm was run for various reservoir parameters, and a relationship between choke size and WOR was observed. Larger water saturation in the stimulated reservoir volume (SRV) delays the onset of hydrocarbon production yet allows choke sizes to be increased at a faster pace. This is attributed to the compressibility of the fluid produced; as incompressible water is produced, pore pressure rapidly declines.


The selection of a choke-management strategy depends on various factors such as the separator pressure or the water saturation in the SRV. Consequently, previous experience and general guidelines may not guarantee a successful production increase and should not be used across the board. A simple application in unconventional oil wells showed that choke sizes can be increased aggressively until the onset of hydrocarbon production if, for example, proppant flowback is an issue of concern. This algorithm can be deployed with a history-matching scheme that uses real-time sensor data to assess reservoir properties better and improve recommendations for future choke adjustments.

This article, written by JPT Technology Editor Chris Carpenter, contains highlights of paper SPE 187190, “Choke Management Under Wellbore, Completion, and Reservoir Constraints,” by Emmanouil Karantinos, SPE, and Mukul M. Sharma, SPE, The University of Texas at Austin, prepared for the 2017 SPE Annual Technical Conference and Exhibition, San Antonio, Texas, USA, 9–11 October. The paper has not been peer reviewed.

Model Helps Guide Choke-Management Strategies Under Constraints

01 September 2018

Volume: 70 | Issue: 9


Don't miss out on the latest technology delivered to your email weekly.  Sign up for the JPT newsletter.  If you are not logged in, you will receive a confirmation email that you will need to click on to confirm you want to receive the newsletter.