Optimization of Steam and Solvent Injection as an Improved Oil Recovery Method for Heavy Oil Reservoirs

Optimization of Steam and Solvent Injection as an Improved Oil Recovery Method for Heavy Oil Reservoirs

10th International Symposium on Process Systems Engineering - PSE2009 Rita Maria de Brito Alves, Claudio Augusto Oller do Nascimento and Evaristo Chal...

189KB Sizes 18 Downloads 430 Views

10th International Symposium on Process Systems Engineering - PSE2009 Rita Maria de Brito Alves, Claudio Augusto Oller do Nascimento and Evaristo Chalbaud Biscaia Jr. (Editors) © 2009 Elsevier B.V. All rights reserved.

1863

Optimization of Steam and Solvent Injection as an Improved Oil Recovery Method for Heavy Oil Reservoirs Edney R. V. P. Galvão,a Marcos A. F. Rodrigues,a Jennys L. M. Barillas,a Tarcilio V. Dutra Jr.,a,b Wilson da Mataa,c a

Programa de Pós-Graduação em Ciência e Engenharia de Petróleo, Universidade Federal do Rio Grande do Norte - Campus Universitário, Natal-RN 59078-970, Brasil b Departamento de Engenharia Química, Universidade Federal do Rio Grande do Norte - Campus Universitário, Natal-RN 59078-970, Brasil c Departamento de Engenharia Elétrica, Universidade Federal do Rio Grande do Norte - Campus Universitário, Natal-RN 59078-970, Brasil

Abstract Currently a resource more and more used by the petroleum industry to increase the efficiency of steam flood mechanism is the addition of solvents. The process can be understood as a combination of a thermal method (steam injection) with a miscible method (solvent injection), promoting, thus, reduction of oil viscosity and interfacial tensions between injected fluid and oil. Solvents are hydrocarbons well known for reducing these tensions and facilitating the production of heavy oil. The use of solvent alone tends to be limited because of its high cost. When co-injected with steam, the vaporized solvent condenses in the cooler regions of the reservoir and mixes with the oil, creating a zone of low viscosity between steam and heavy oil. Mobility of the displaced fluid is then improved, resulting in an increase of oil recovery. To better understand this improved oil recovery method, a numerical study of the process was done contemplating the effects of some operational parameters (distance between wells, steam injection rate, solvent type and injected solvent volume) on cumulative oil production and oil rates. A semi synthetic model was used. Some reservoir data were obtained similar to those found in Brazilian Potiguar Basin and others ones were obtained from literature. Simulations were performed in STARS (CMG, 2007.11). It was found that injected solvent volumes increased oil recovery and oil rates. Further, the majority of the injected solvent was produced and can be recycled. High initial productions achieved by models that use solvent have normally a significant impact on the operation economics, because earlier productions suggest that fluids injection (steam and solvent) can be interrupted earlier. On environmental point of view, it would have a reduction of energy and water consumptions for steam generation, having diminished Green House Gases (GHG) emissions. Also it is important to emphasize that the high oil rates presented by these models can generate an earlier financial return, and this would be decisive for the economic viability of the project. Keywords: solvent, steam flood, heavy oil, reservoirs modeling.

1. Introduction Steam flood is an improved oil recovery method applied generally in viscous oil reservoirs. This method consists of injecting heat to reduce viscosity, increasing oil

1864

E.R.V.P. Galvão et al.

mobility and facilitating its production. To increase the efficiency of this mechanism, a resource more and more used by the petroleum industry is the addition of solvents. Diverse hybrid steam-solvent processes are being developed, mainly by Alberta Research Council (ARC), in Canada (Nasr & Ayodele, 2006). Experimental results, numerical studies and field tests suggest that great benefits can be achieved with the addition of solvents to the injected steam, as an increase of oil rates and oil-steam ratios, reduction of energy and also a reduction in water consumptions for steam generation, having diminished Green House Gases (GHG) emissions. Thus, it is indispensable to study the viability of this new technology in heavy oil fields from Brazil, with right adjusts to particularities of each reservoir.

2. Process Modeling In this work, steam and solvent injection was analyzed through a vertical wells system. A semi synthetic model was used. Some reservoir data were obtained similar to those found in Brazilian Potiguar Basin and others ones were obtained from literature. Simulations were performed in STARS (Steam, Thermal and Advanced Reservoir Simulator) 2007.11 from CMG (Computer Modelling Group). 2.1. Reservoir Modeling The physical model corresponds to an oil reservoir of 100 m x 100 m x 29 m, on a Cartesian coordinates system (x, y and z directions), as shown in Fig. 1. To reduce simulation time and considering system symmetry, the chosen flooding pattern was one quarter of an inverted five-spot, represented by an injector and a producer wells. Some of used data, like rock-fluid properties and heat loss parameters, are represented in Table 1. In the bottom hole direction, injector well was completed until the 8th layer, while producer well was completed until the 11th one.

Fig. 1. Reservoir model. Initial oil saturation. Table 1. Reservoir simulation input parameters.

Item Reservoir thickness (m) Oil zone thickness (m) Water zone thickness (m) Initial reservoir temperature (ºC) Reservoir depth (m) Porosity (%) Vertical/horizontal permeabilities ratio (%)

Value 29 20 9 37.8 200 30 10

Optimization of Steam and Solvent Injection as an Improved Oil Recovery Method for Heavy Oil Reservoirs

1865

Table 1. Reservoir simulation input parameters (continuation).

Item Oil viscosity* Initial oil saturation (%) Irreducible water saturation (%) Steam temperature (ºC) Steam quality (%) Maximum pressure in the injector well (kPa) Minimum pressure in the producer well (kPa) Thermal conductivity of rock and surrounding formation overburden and underburden (J/m-s-K) Thermal conductivity of water (J/m-s-K) Thermal conductivity of oil (J/m-s-K) Thermal conductivity of gas (J/m-s-K) Volumetric heat capacity of surrounding formation overburden and underburden (J/m3-K) Production time (yr)

Value 1000 [email protected]°C 70 30 288 55 7,200 197 1.7 0.6 0.13 0.04 66,465 16

*Viscosity versus temperature curve was obtained from Barillas, 2005.

Equations from mathematical modeling are embedded in STARS, which is a commercial software. Thus, it is not necessary to insert them in the input data file. Description of the process is done using specific keys from STARS library. Input parameters depend on available rock-fluid properties, as ones showed in Table 1. Because ideal models were used, they were not compared with real models. 2.2. Analyzed Operational Parameters The minima (-1), intermediates (0) and maxima (+1) levels as well as nomenclature of analyzed operational parameters are shown in Table 2. A factorial planning of three levels was realized to study the effects of parameters interactions, resulting in 81 simulations. Table 2. Range of analyzed operational parameters.

Parameter Distance between injector/producer wells (m) Steam injection rate (m3/day) Solvent type Injected solvent volume/Injected steam volume (%)

Minimum (-1)

Intermediate (0)

Maximum (+1)

65

100

135

20 Pentane

35 Hexane

50 Heptane

5

10

15

Used maximum production rates correspond to the double of respective values of fluids (steam + solvent) injection rates. Oil recoveries (OR) were calculated dividing cumulative oil (CO) by original oil in place (OOIP) associated to each grid configuration (OR = CO/OOIP). Being a function of rock total volume, porous volume also varied, while all the others reservoir properties remained constant.

1866

E.R.V.P. Galvão et al.

3. Results Through the analysis of parameters interactions and taking oil recovery as objective function, an optimized model was achieved. Table 3 shows the parameters from this model and others models used to emphasize the effect from injected solvent volume and to make a comparison with the case in which steam is injected without solvent. Table 3. Operational parameters values for models 0%, 5%, 10% and 15% of injected solvent.

Parameter

0% Solvent

5% Solvent

10% Solvent

15% Solvent (Optimized Model)

Distance between injector/producer wells (m)

100

100

100

100

Steam injection rate (m3/day)

20

20

20

20

Solvent type Injected solvent volume/Injected steam volume (%)

_

Heptane

Heptane

Heptane

0

5

10

15

3.1. Comparison among Primary Recovery and 0%, 5%, 10% and 15% Solvent Models In Fig. 2, cumulative oil (without solvent) versus time are showed for primary recovery and models listed in Table 3. From second year on, curves with solvent models registered cumulative oil (without solvent) higher than the ones obtained by model 0% Solvent. This suggests that presence of solvent, for the analyzed amounts, accelerated the arrival of warm oil as far as producer well, promoting, thus, an earlier production with regard to model 0% Solvent. Moreover, it can be evidenced that an increase of injected solvent volume improves oil recovery (70,3% for model 5% Solvent, 71,1% for model 10% Solvent and 72,4% for model 15% Solvent).

Fig. 2. Cumulative oil (without solvent) versus time. Primary recovery and models 0%, 5%, 10% and 15% Solvent.

Optimization of Steam and Solvent Injection as an Improved Oil Recovery Method for Heavy Oil Reservoirs

1867

Fig. 3 shows oil component (without solvent) rates versus time for these models. It can be observed that the higher injected solvent volume, the earlier production rates. This effect occurred mainly between second and fourth year, in which Optimized Model (15% Solvent) achieved 15.0 m3/day, while model 0% Solvent achieved only 5.0 m3/day at the same time. High rates obtained initially by Optimized Model contributed for a faster reservoir depletion, justifying the lower rates achieved by one in the following period.

Fig. 3. Oil component (without solvent) rates versus time. Primary recovery and models 0%, 5%, 10% and 15% Solvent.

High initial productions achieved by models that use solvent have normally a significant impact on the operation economics, because early production suggests that fluids injection (steam and solvent) can be interrupted earlier. On environmental point of view, solvent injection can provide a reduction of energy and also a reduction in water consumptions for steam generation, having diminished Green House Gases (GHG) emissions. Also it is important to emphasize that the higher oil rates presented by these models can generate an earlier financial return and, by consequence, a project with a good economical viability (Galvão, 2008). Injected and produced volumes of solvent versus time for models are registered in Fig. 4. Practically all injected solvent was produced together with oil from reservoir. To quantify the reuse of produced solvent is not in the scope of this work, however it is expected that produced fluid is lighter (high ºAPI) and has higher economic value (Galvão, 2008). Also it can be evidenced that the higher injected solvent volume, the earlier its recovery began.

1868

E.R.V.P. Galvão et al.

Fig. 4. Injected and produced volumes of solvent versus time. Models 5%, 10% and 15% Solvent.

4. Conclusions • An increase of injected solvent volume improved the cumulative oil production with regard to model without solvent; • Presence of solvent, for the analyzed amounts, accelerated the arrival of warm oil as far as producer well, promoting, thus, an earlier production with regard to model without solvent. This suggests that fluids injection (steam and solvent) can be interrupted earlier, reducing energy and water consumptions for steam generation, Green House Gases (GHG) emissions and generating an earlier financial return; • Practically all injected solvent was produced together with oil from reservoir. It is expected that produced fluid is lighter (high ºAPI) and has higher economic value. Also it could be evidenced that the higher injected solvent volume, the earlier its recovery began.

Acknowledgements The authors want to thank CAPES and Laboratório de Estudos Avançados em Petróleo (LEAP-UFRN) for the support received in the execution of this work.

References NASR, T. N.; AYODELE, O. R. New hybrid steam-solvent processes for the recovery of heavy oil and bitumen. SPE 101717. Nov, 2006. GALVÃO, E. R. V. P. Injeção de vapor e solvente como um método de recuperação avançada em reservatórios de óleo pesado. 2008. 106p. Dissertation (Master Degree in Petroleum Engineering) – Centro de Tecnologia, Programa de Pós-Graduação em Ciência e Engenharia de Petróleo, Universidade Federal do Rio Grande do Norte, Natal. CMG, Computer Modelling Group Ltda. Guía para el usuario. Steam, Thermal and Advanced Reservoir Simulator - STARS. Versão 2007.11, Calgary-Alberta-Canada. BARILLAS, J. L. M. Estudo do processo de drenagem gravitacional de óleo com injeção contínua de vapor em poços horizontais. 2005. 163p. Dissertation (Master Degree in Chemistry Engineering) – Centro de Tecnologia, Departamento de Engenharia Química, Programa de Pós-Graduação em Engenharia Química, Universidade Federal do Rio Grande do Norte, Natal.