Transient Simulation for Hydrate Mitigation Aims for Optimal Production in Kuwait Fields

Topics: Flow assurance
Fig. 1—Hydrate curve for well producing condensed water without MeOH injection.

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

In the deep high-pressure/high-temperature North Kuwait Jurassic (NKJ) fields, the pipelines connecting the wells to the processing facility are neither buried nor insulated. During the winter, the well fluid cools to below hydrate-formation temperature in the flowline, causing hydrate crystallization and even plugging. This paper presents the traditional methods of hydrate mitigation used in the NKJ fields and the way in which a transient model was initially built and continuously improved.

Challenges

Hydrate Formation. When the well forms hydrates, usually at night and early morning in winter, the field operators must wait for the ambient temperature to rise in order to melt the hydrate plugs. In general, well production declines for 6–8 hours because of hydrates. Because hydrate formation is the main cause of concern, a robust solution is needed to minimize production downtime. No proper flow-assurance study or modeling was conducted to understand the effect of water and inhibition details on the hydrate curve. To address this challenge, a predictive transient tool is needed to know in advance when hydrate will start to form, the location of hydrate formation, and the required methanol (MeOH) dosage rate to avoid hydrate formation. The current practice of hydrate inhibition is mainly based on past experience. Injecting MeOH on the basis of past experience, without knowing the optimal MeOH-injection rate, could lead to other flow-assurance challenges.

Slug Flow. Low-condensate/gas--ratio and high-water/gas-ratio wells have a tendency to create slug flow in the pipeline because of continuous changes in the phase holdups. Because of wide temperature fluctuations between night and day, the fluid in the pipeline is always in a transient state. The pipeline acts as storage because of its large volume (1,000–3,000 bbl). In the morning, when the pipeline is heated, the -fluids expand and gas holdup increases in the line, which pushes the liquid to the facility and causes a surge at the facility inlet. At night, when the pipeline cools, liquid hydrocarbon accumulates in the line, increasing the liquid holdup. High-water wells always will have a tendency to accumulate water in the low-lying zones. This water generally moves as a slug. Modeling of the multiphase fluids to understand the slug behavior is required for balancing the fluid in and fluid out of the facility. Long and complicated flowline geometry further complicates the system dynamics and will affect the flow behavior.  

Pipeline Corrosion. An acid-gas environment with water production makes the system very corrosive. The current practice is to inject 10 L/h of corrosion inhibitor, which is expected to coat the whole pipeline and protect it from corrosion. 

Hydrate-Inhibition Operational Envelope. The available pumps have a maximum capacity of 20 L/h per pump, which is not enough. Often, two or three pumps are installed and connected to a single injection port at the wellhead to pump 40–60 L/h, which is not a proper solution and is not foolproof. Currently, the MeOH injection is performed manually and pumping is continued throughout the day during the winter.  

The injection pumps and the chemical tanks are not connected to the supervisory control and data acquisition system; therefore, the monitoring of the pumps and chemical consumption does not happen in real time. Pump failure cannot be detected remotely; thus, routine checkups are required.

Current Well Modeling. The current well and pipeline models are steady state and are not capable of simulating the real dynamic behavior that causes hydrate and slug flow.

Liquid Loading. Automated chokes are used for hydrate control. Choke reduction will result in liquid loading in some of the low-reservoir-pressure wells, which further reduces well productivity. Proper design of hydrate inhibition will eliminate the need for the choke reduction and, thus, avoid liquid loading. 

Solution

To address some of the challenges, a transient modeling solution began in 2014 for the producing NKJ wells. In the first phase, a pilot of an offline transient solution was deployed, with hydrate and surge advisers as the key deliverable. Massive data gathering was conducted from different disciplines to create a robust transient solution.

Hydrate Adviser. From the initial -hydrate-prediction results, additional water-sample analysis was conducted to account for different salts in the formation water. On the basis of the transient-modeling and sampling results, NKJ wells can be divided into three categories. 

Category A Wells: Wells That Produce No Formation Water. These wells have no brine production because water present in the gas condenses when the fluids reach the surface during flow. The water content in these wells is so low (<5%) that it falls out of the water-measurement range of available multiphase flowmeters. The modeling of Category A wells was performed by saturating the fluid composition to the reservoir pressure and temperature to account for the condensed water from the gas phase. A hydrate curve is generated for the case of no hydrate inhibition. Fig. 1 above shows the modeling result—which matches actual field observation—where hydrate starts to form when the ambient temperature drops below 20°C.

To model the effect of hydrate-inhibitor injection, the same well was modeled with a 40-L/h injection rate, as shown in Fig. 2. The hydrate curve shifts to the left, which will result in hydrate formation at lower ambient temperature. This also has a fair match with the actual field data. 

Fig. 2—Hydrate curve for well producing condensed water with MeOH injection.

 

Category B Wells: Wells That Produce Saline Formation Water. For these wells, a more detailed overview of the concentrations of different types of salt is required to reduce the uncertainty of the hydrate-formation temperature predicted by the model. Sensitivity analysis was conducted for different salt types and their effect on the hydrate curve. Because different salts have different inhibition effects, having the actual salt distributions is important to achieve a realistic hydrate curve. The hydrate curves with detailed salt descriptions have a close match with actual field conditions. 

Category C Wells: Wells That Produce High-Salinity Formation Water.  Paying special attention to the wells that are producing formation water with high salt concentrations (>280,000 ppm) is important. Overdosing with MeOH is not recommended for high-salinity-water wells because of the “supersaturated salt” or “salt-out” effect.

Surge Adviser. The surge adviser is part of the transient modeling solution and is important for accounting for changes in the flow behavior in the pipeline resulting from various factors. The surge adviser predicts the surge at the inlet of the facility and allows production balancing to be accomplished. 

Corrosion and Gas Velocity. In general, the fluid velocity in gas pipelines should be less than 60 to 80 ft/sec to minimize noise and to allow for corrosion inhibition. A lower velocity limit of 50 ft/sec should be used in the presence of known corrosive agents such as carbon dioxide. The minimum gas velocity should be between 10 and 15 ft/sec, which minimizes liquid fallout. If the velocity is high, the fluid will disturb the corrosion-inhibitor coating and generate corrosion sites. 

Conclusion

Hydrate Formation. Transient modeling can mimic field conditions if accurate water properties are used to generate the hydrate curve. The effect of the salts present in the formation water and the MeOH injection is well-understood, and MeOH injection can be optimized by the use of transient modeling. The solution should be scaled up to include all the new wells, and the system should be used to minimize the production losses resulting from hydrates.

Slug Flow. The sensitivity analysis conducted for ambient temperature and the choke changes clearly demonstrates that the flow from wells will always have slug behavior at the inlet of the facility. The long pipelines of the NKJ fields act as fluid storage, and their geometry adds to the slug behavior. When the system is on line, the slug adviser can be used to optimize fluid flow in and out of the early-production facility.

Transient Modeling. The steady-state models are capable of simulating flow-assurance scenarios. Dynamic hydrate and surge advisers can be used with the transient modeling. The key function of the hydrate adviser is to monitor the temperature margins of potential hydrate-formation regions throughout the production network. The hydrate margin calculation takes into account actual local pressure, temperature, and amount of inhibitor.

Similarly, the key function of the surge adviser is to calculate the formation, location, and size of terrain slugs or liquid surges in the production system. The surge adviser provides a forecast of the effect of upcoming slugs/surges on the early-production-facility high-pressure-separator conditions (liquid levels and pressure) and provides advanced warning if the alarm limits are exceeded.

Pipeline Corrosion. Modeling of the gas velocity in the pipeline is important for predicting pipeline corrosion. This study will be expanded in the future to include the effects of the type and rate of corrosion inhibition. 

This article, written by Special Publications Editor Adam Wilson, contains highlights of paper SPE 182237, “Assuring Optimal Production and Enhanced Operational Efficiency Through Transient Simulation—A Case Study in North Kuwait Jurassic Fields,” by Mohammad Al-Sharrad, Kuwait Oil Company; Roshan Prakash, SPE, and Christian F. Trudvang, SPE, Schlumberger; and Noura Al-Mai, Abrar A. Hajjeyah, SPE, and Abdulaziz H. Al-Failakawi, SPE, Kuwait Oil Company, prepared for the 2016 SPE Asia Pacific Oil and Gas Conference and Exhibition, Perth, Australia, 25–27 October. The paper has not been peer reviewed.

Transient Simulation for Hydrate Mitigation Aims for Optimal Production in Kuwait Fields

01 November 2017

Volume: 69 | Issue: 11

STAY CONNECTED

Don't miss the latest content delivered to your email box weekly. Sign up for the JPT newsletter.