Hybrid System of ESP and Gas Lift Application from Conceptual Design, Pilot Test to System Analysis

Hybrid System of ESP and Gas Lift Application from Conceptual Design, Pilot Test to System Analysis

Artificial lift technology application in heavy oil production has been far-reaching developed in the industry over past decades by persistent efforts to improve the ultimate recovery of this “difficult” hydrocarbon. Heavy oil discovery in a marginal field, Cuu Long Basin, Offshore Vietnam is relatively aberrant and pose challenges to full field development. A series of systematic technical studies have been purposely planned from the first discovery of heavy oil in the wildcat well to the modeling study and facility design to accommodate the viscous fluid over the field life. Apart from the thermal method, pumping technology makes remarkable advance by enlarging the drawdown created over the conventional gas lift in several heavy oil projects. After due consideration, the Electrical Submersible Pump (ESP) was finally decided as the key driver to reinforce well production performance. Moreover, the gas lift has been brought in as a backup in case of pump failure which is not only to prolong well life, save workover expenditure but also boost production if operating in hybrid mode.

This paper presents sequential events from the conceptual study to pilot test hybrid ESP/Gas lift system and ultimately the inflow/outflow curves analysis. A proper system analysis of the inflow/outflow curves is indispensable to model the outflow curve above the pump where the aid of gas lift complicated the upward flow and to generate the lift curves used in reservoir simulation. The pilot test of this electro-gas system to Well A has shown about 30% liquid production increment with lesser pump energy consumed and flexibility in control operating point. The early results promise further extension to the remaining ESP wells to enhance field production.

Introduction

The combination of Electrical Submersible Pump (ESP) and Gas lift (GL) has been firstly used in a viscous oil in marginal filed, CuuLong Basin, offshore Vietnam. The field was found commercially via three wildcat wells with a wider range of formation and fluid properties. Its development concept is depicted in Figure 1 in which heavy oil producers in Platform B will be the candidate for artificial lift optimization. Three pay zones were discovered in the Middle Miocene Upper-Lower Con Son formations (BII.2.20, BII.2.30, and BII.1.10) in wildcat wells A, B and C as stacked channel sandstones trapped. Each gross sandstone package is about 30-40m thick, capped above by 10-30m of shale/clay stones. Figure 2 illustrates the vertical cross section over the wildcat wells. The diverse in fluid properties (20.5⁰API to 35⁰API) posed difficulty in selecting production technology method to enhance wellbore lifting efficiency.

Figure 1 - Field layout

Figure 2 - Reservoir cross-section over three wildcat wells

A systematic approach to studying all artificial lift methodology has been performed in order to select the most suitable/productive producing method. The main purpose of studying is by looking into allTable 1reservoir conditions and production requirements, and it has come out by following:

•       Maximize the overall lifting efficiency of the well
•    Accelerate production by applying more aggressive pressure drawdown
•    A thorough understanding of the system analysis for future field-wide application

Table 1 - Reservoir conditions and production requirements for pump sizing

Parameter	Value
Reservoir Pressure	1946 – 2,192 psia
Reservoir Temperature	171 – 174, deg. F
Tubing Head Pressure	250 psia
Pump Measured Depth	4,406 – 4,987 ft
Designed Rate	1,500 bbl/day
Gravity	20.50API
Viscosity @ pump depth	8.9 cp
PI	3-4 bbl/psi/d
Total GOR	5-10 scf/STB
Water Cut	0%-90%

A production technology study which imposed multidiscipline data from the reservoir, well to the surface facility, was carried out to best select the reservoir candidate and appropriate artificial lift method in the Field Development Plan. For this particular reservoir, the process is depicted in Figure 3. Few technologies were considered including PCP, EPCP but ESP was the final selection based on its technology advancement well fit the reservoir conditions, development concept, increasing oil production by lowering bottom hole flowing pressure and increasing reserve by lowering abandonment pressure. Reservoir simulation forecasted 5 percent addition in recoverable reserve with the ESP application over GL.

Figure 3 - Artificial lift selection criterion

The study has come out with resultsthatESPsarethe best-suited lifting method for all well conditions and requirements. Compared to traditional and familiar lifting methods, gas lift, in Vietnam oilfield, ESPs can provide higher production rates to compensate for the declining oil production and increasing water cut. More importantly, the dual system ESP and Gas lift is taken into account, as this combination provides a wide operating range, allowing individual operation of the primary, the secondary and the combined system, reaching the optimal technical and economic performance while the combined system is operating. The long-term benefits are a reduction in production downtime and flexible capacity to ramp up production where necessary.

However, there are also some concerns related to the run life of ESP. For instance, the production of solids as is the nature of shallow and unconsolidated reservoirs, if not managed properly, will potentially lead to a reduction in run life. Therefore, a rigorous geometric and sand control study during the completion design phase was performed beforehand to ensure sand-free fluid flow in the wellbore. The other major concern is in the case of an ESP failure, it will cause a large drop in production until the ESP is retrieved and replaced. In this case, the GL systems backup will play a key role to partially compensate for this production loss during the time waiting for the ESP replacement.

Conceptual Design

Figure 4 shows the equipment string where GL system is placed above ESP packer to provide a backup system as well as optimization capability in case of simultaneously producing by both ESP and GL. The design and selection of ESP system are evaluated by taking the well test and potential change of reservoir parameter into consideration. With respect to the ESP, Figure 5 shows pump VSD curve per stage of the selected pump, the compression pump with a suitable operating range to achieve the target production rate (1,500 bpd) was selected to efficiently drive the viscous fluid up to the surface by using the benefits of a mixed flow stage design. With the candidate well-having low GOR (approx. 10scf/stb) and the intake pressure is far from bubble point pressure (approx. 100psi), a gas separator is not required for this case and a conventional intake with the pump is sufficient to work well with zero percent free gas into the pump. Two tandem L/2BP type seals were selected providing sufficient chamber expansion capability and thrust load capacity. A motor with an ESP downhole sensors provided real-time-view of ESP performance as well as the downhole operating conditions. Corrosive resistant housing material was used to for the ESP equipment to better withstand the corrosive environment. Surface equipment; which is also significant system to control ESP downhole system;Active Front End (AFE); Variable Speed Drive (VSD) was used to be key controller offering the capability to optimize the production by adjusting ESP operating frequency in order to obtain optimum production with a smaller footprint required on the offshore platform than a conventional VSD. The pump sizing resulted in the number of stages varied either 37ea or 60ea, depending on reservoir conditions.

Figure 4 - A typical Dong Do ESP well completion schematic

Figure 5–Selected pump VSD curve per stage

Installation phase

From the day of finalizing equipment to meet all requirements until the readiness of equipment in the operational facility, the transition to the operational phase, the execution plan (as described in Figure 5Operational execution process) was made in order to shape and secure all on-going process in line with schedule and target. Surface equipment was first installed to wellhead platform and pre-commissioning of equipment was also performed to ensure equipment functioned as it was designed. Downhole equipment was systematically tested with other related equipment making sure all components properly fit with each other without any sign of potential risk and failure during installation and operation. Equally, important was installing all combined system components into the well with a proper and safe manner as planned. Starting up the well to achieve production within an optimized condition and designed operating range.

Figure 6 – Operational execution process

Five (5)wells have been successfully installed with ESP downhole equipment as shown in Table 2 – Installation records in which all wells have been efficiently operating within pump recommended operating range in order to prolong run life and performance. Some of well have been simultaneously operating ESP and GL and result found is the more of production gained and also reduction of Horsepower (HP) consumed.

Table 2 – ESP installation records

Well	Installation date	AL method
Well #1	May 2015	Electro-gas
Well #2	June 2015	Electro-gas
Well #3	June 2015	Electro-gas
Well #4	June 2014	ESP
Well #5	August 2014	ESP

Pilot Test

The pilot test was conducted successfully on three wells since May 2015. Both wells underwent instantly incremental oil gained to date. For the purpose of examining the hybrid ESP/GL application, well #3 is selected and denoted as well A through the succeeding sections. Figure 7 illustrates the production performance of well Aat the ESP only and hybrid mode. With identical system settings, it is apparent that liquid rate increased from 1,340 bpd to around 1,740 bpd or 30 percent of liquid increment.

Figure 7 - Production performance of well A

Well A was commenced on 3rd July 2015 and on ESP mode at a frequency of 53 Hz. The initial rate achieved 1,610 bopd with no water production. However, the first trace of produced water was recorded after a two-week on stream and water cut quickly ramped up to 40% within a month. ESP frequency was then reduced to 49Hz in an attempt to delay the increasing trend of water cut. Water cut then stabilized at around 50% while oil rate gradually dropped to 720 bopd. In mid of December, the well switched to the electro-gas mode by injecting gas lift at 0.2 MMscf/d in conjunction with ESP frequency of 49Hz. As electro-gas was applied, well A performance showed improvement in term of incremental liquid of 30 percent.

It is apparent that the hybrid system at this specific condition is somehow comparable to a single ESP running at 55 Hz. This fundamental finding is intriguing and important for reservoir engineer to make the Vertical Lift Performance curves for reservoir simulation since no commercial software is capable of modeling the hybrid ESP/GL application.

System Analysis

The gradient transverse plot is useful to address pressure distribution along the tubing string and to spot out the slope change at specific depth coincident with gas lift valves or pump depth. Examining the gradient transverse plot for a well fitted with a hybrid ESP/GL system, it can be seen that there are four points that characterize the shape of the curve; tubing head pressure, gas lift injection, discharge pressure, and intake pressure. In addition, the reservoir pressure is obtained via the pressure sensors mounted at the bottom part of the pump. The relationship between the parameters can be defined as follows.

Above the pump

P_d = THP + ΔP_g1 + ΔP_f1 + ΔP_a1

And below the pump

P_wf = P_i + ΔP_g2 + ΔP_f2 + ΔP_a2

Where (1) and (2) denoted for above and below the pump

As mentioned in the previous section, there are three wells currently operating in electro-gas mode and it is really necessary to be able to model these wells with the aim of potential production optimization during the well life. There are many well-known software tools in the industry that are marketed as being capable of performing ESP analysis and diagnosis. Unfortunately, most of them are virtually unable to model a dual artificially lifted well, in this case, the ESP and GL combination. However, this challenge can be overcome by constructing both artificial lift models separately, using the results of each to find a solution for both as illustrating in Figure 8. One typical electro-gas well is modeled in two separate sections: ESP and GL, where the output of the ESP section is the input of the GL section.

­   ESP section: this section is built with the ESP, tubing, and constructed only from reservoir to the pump setting depth. After successfully modeling the well, well test data is used to perform matching. 

 GL section: this section employs GL system with the reservoir being the output of the ESP section. The model was built from the pump setting depth up to the wellhead with existing GL design. The model then verified by matching with production data

The workflow presented in Figure 9 lists the breakdown steps in matching individual ESP/GL Section and finally the hybrid one against flow test data.

Figure 8 -Hybrid ESP/GL model concept

Figure 9 – Hybrid ESP/GL calibration process

For the purpose of system analysis, flow test was conducted in the period from February to March 2016 where well A flowed stable at Frequency 49 Hz and water cut around 72 percent. Three flow test points were examined systematically with (A) only ESP, (B) ESP and GL 0.2 MMscfd and (C) ESP and GL 0.28 MMscf as illustrated in Table 4. It should be noted that 0.28 MMscfd is the max gas injection over orifice 1/8” initially designed to unload the well when the ESP is down. The incremental liquid between Test A and Test B is quite considerable around 30 percent and continues rising with more gas injected.In similar, the power consumption also reduces slightly 2 percent as a result of a lighter fluid column above pump created by gas lift injection.

Table 4 - Summary well test points for system analysis

	Test A	Test B	Test C
Liquid rates (stb/d)	1,340	1,740	1830
Water cut (%)	72	72	72
WHP (psig)	294	441	448
Gas lift rate (MMscfd)	0	0.2	0.28
Pd (psia)	2,140	2,060	2,052
Pi (psia)	1,670	1,650	1,664
Frequency (Hz)	49	49	49
HP (KVA)	31.8	31.2	31

In the ESP model section, the IPR curves at the reservoir depth and at the pump setting depth are shown in Figure 10 below. The real well-test data are used in the matching process to ensure the model reflecting the real condition. In this case, the ESP pump intake and discharge pressure at the setting depth, along with production rate in the model coincided with well test date.

The ΔP across the pump is affected by frequency, flow rates, the number of stages, fluid properties, and pump efficiency. The ΔP value is converted to head and plotted on the pump curve. As a result of reducing density above the pump, the pairs of intake/discharge shifted downwards from 1,670 psia/2,140 psia with only ESP to 1,650 psia/2,060 psia in the hybrid ESP/GL 0.2 MMscfd. This is feasible by tuning the PI slightly higher (4.5 stbd/psi) than the original input. Note that the pump curve and head per stage would change considerably to accommodate the declines in pump intake/discharge, the ΔP, and increasing flow rates even though the frequency is locked at 49 Hz. During the trial period, sometimes the ESP system operating point dragged to the up-thrust region and out of the designed pump curve. Hence the recalibration of the pump curve is indispensable.

Figure 10 -The pressure transverse curve of the ESP Section

After having the output data, flow rate and pump discharge pressure will then be used as input in the inflow section in the GL model. In gas lift section, the model is built from pump setting depth up to the wellhead, with the inflow model using input data from ESP model including pump discharge pressure and production rates. The model is matched using well test data in electro-gas phase and illustrates the pressure transverse plot in Figure 11. Since the PVT is assumed to be unchanged, the lesser fluid density forced the pump discharge to reduce 80 psia while THP increases 154 psi (from 294 psia to 448 psia). There is a modest rise in pressure and flow rates when injecting more gas.

Figure 11 -The pressure transverse curve of the GL Section

After completing both models for ESP and GL, the final step is to couple them into one model to identify system response in a certain trend that could be further extended to higher gas lift rates and applicable to another well. The analysis is also valuable for production optimization, production forecast, and reservoir simulation. The system analysis of the testing period is detailed in Figure 12 where sets of inflow curves extracted from the ESP section model and the outflow from the GL section model.

With increasing gas lift injection, the system responded comparably to the increase of the frequency of the system with ESP and no GL in terms of incremental liquid rates but contrary in the ΔP. Let’s look back to Figure 7 in the period July to August 2015 in which well A was operated with solely the ESP. The well had been producing at a liquid rate 1,730 bpd but with the ESP frequency up to 55 Hz. Apparently, the pump had to work harder, consumed more power to deliver such flow rates. The inflow/outflow response of the hybrid ESP/GL system is favorable to the ESP run life in the sense of making lessΔP across the pump, 470 psia at no GL down to 388 psia at GL 0.28 MMscfd, but still able to lift more. Mathematically, with that three datasets, it is possible to predict the system response when increasing frequency and/or gas lift rates. This way we may even run the pump with exceeding its max frequency 58 Hz without replacing the new pump, lower bottom hole pressure and considerably improve well deliverability.

Figure 12 -System analysis of the hybrid ESP/GL system

Conclusions

Marginal fields require cost-effective and high-efficiency concepts to maximize revenues and make the project economics. The pathway to the installation of the ESP this field is challenging but rewarding as it is the first ever application successful in Vietnam and brings enormous benefits for the company.

·         Prudent studying reservoir performance is essential to understand the specifics of the flow dynamics to assist in concept selection.

·         Dual artificial lift systems help reducing production downtime and provide the flexible capacity to ramp up production when necessary.

·         Injecting gas lift lightens the fluid column above the pump making changes in the fluid properties, leading to less ΔP across the pump and enhances pump run life.

·         The hybrid application in the testing period results in 30 percent liquid rate increment while the power energy required is 2 percent less than solely ESP when injecting gas lift at the rate of 0.28 MMscfd.

·         The electro-gas system allows the pump operated beyond its designed frequency/flow rates which ultimately result in a cost saving of replacing the new pump and improve oil recovery factor.

Acknowledgment

The authors gratefully acknowledge Lamson Joint Operating Company, GE Oil and Gas Inc. and Eastsea Star Software Company for various supports and permission to publish this work.

Nomenclature

PI = Productivity Index (bpd/psi)

P_d = discharge pressure (psi)

P_i = intake pressure (psi)

L/2BP = Labyrinth series Two Paralleled Bags 

THP = tubinghead pressure (psi)

ΔP_g = hydrostatic head (gravity pressure loss in psi)

ΔP_f = frictional pressure loss (psi)

ΔP_a = acceleration pressure loss (psi)

P_wf = bottomhole flowing pressure (psi)

P_r = reservoir pressure (psi)

Q = well flowrate in stock tank barrels liquid (stbl/day)

PCP = Progressing Cavity Pump

EPCP = Electric Progressive Cavity Pump

VSD = Variable Speed Drive

AFE = Active Front End

References

Detailed Well Completion Design, Lamson Joint Operating Company - Internal Report, 2012.  

Key, M., Heath, R., Web-Ware, A., High Horsepower ESP Application Challenges in Offshore Marginal Fields, SPE MEALF, 2008.

Bokor Field ESP Conversion - Completion Selection Report, 2007.

Qahtani, A.A., Electric Submersible Pump (ESP) Selection Optimization: A Reservoir Engineering Outlook, MEALF,2007.

A. J. (Sandy) Williams, “Demystifying ESPs: A Technique to Make Your ESP Talk to You”, ESP Workshop Houston, 2000. 

This paper was prepared for presentation at the SPE Middle East Artificial Lift Conference and Exhibition held in Manama, Kingdom of Bahrain, 30 November–1December 2016
Link: https://www.onepetro.org/conference-paper/SPE-184215-MS