Numerical Modeling of ExxonMobil s Electrofrac TM Field Experiment - PowerPoint PPT Presentation

Numerical Modeling of ExxonMobil ’ s Electrofrac TM Field Experiment at Colony Mine Nazish Hoda, Chen Fang, Michael W. Lin, William A. Symington, and Matthew T. Stone ExxonMobil Upstream Research 30 th Oil Shale Symposium October 19, 2010

ss Ele Electr trofr ofrac TM TM Pr Proc ocess Oil Sha Oil Shale le C Con onversion via sion via Ele Electric trically C lly Conduc onductiv tive F Fractur tures s • Oil shale is heated in situ by a hydraulic Ele Electr trofr ofrac Pr Proc ocess Sc ss Sche hematic tic fracture filled with an electrically conductive material 1 • Electricity is conducted from one end of the fracture to the other, making it a resistive heating element • Heat is conducted into the formation, converting the kerogen into oil and gas • Oil and gas are produced by conventional methods • Potential for cost-effective recovery with less surface disturbance than: • Mining and retorting • Competitive in situ processes • Several years of research are required to demonstrate technical, environmental, and economic feasibility 1 U.S. patent 7,331,385 B2 2

Location of Loc tion of Exx ExxonMobil onMobil’s C s Colony Mine olony Mine Mine Mine Benc nch h Parachute hute Battle ttlement Me nt Mesa sa Tunne unnels ls I-70 I-7 3

Building a uilding and Ope nd Operating ting Ele Electr trofr ofracs Built and verified Built and verified Decline Decline two Electrofracs : two Electrofracs : • Pumped two Electrofracs with EF1 ~140 ft EF1 ~140 ft EF3 ~200 ft EF3 ~200 ft calcined coke. Mapped and verified with observation Crosscut Crosscut EF3 EF3 132 ft 132 ft coring. Instrumentation Instrumentation Holes on EF1 Holes on EF1 N N o o r r t t h h • Pumped/squeezed Electrofrac D D r r i i f f t t power connections at EF1 and EF3. Preceded by lab and field S S 99 ft 99 ft o o u u t t Cored 28 observation Cored 28 observation h h D D pretests. holes: All intersections holes: All intersections r r i i f f t t EF1 EF1 probed are electrically probed are electrically connected connected Connection Connection Treatments Treatments • Instrumented and heated C C o o k k e e - - f f i i l l l l e e d d E E l l e e c c t t r r Electrofracs in low-temperature o o f f r r a a c c tests M M E E i i n n l l e e e e • EF3 EF3 in 2 in 2009 c c t t R R r r i i c c i i b b c c u u r r r r e e n n t t • EF1 EF1 in 2 in 2010. . Insulated Coating on Insulated Coating on Steel Pipes Steel Pipes 4

tion EF1 & EF1 & EF3 EF3 Instr Instrum umenta ntation Captur ptures T s Tempe peratur ture, V , Volta oltage, C , Cur urrent a nt and R nd Roc ock Mo Movement D nt Data ta 16 F Fibe iber Optic r Optics H s Hole oles s Nor orth th 16 T The herm rmoc ocouple ouple H Hole oles s 11 T The herm rmoc ocouple ouple / / osscut ut Fibe iber H r Hole oles s Crossc EF3 EF3 EF1 EF1 5

Much D Muc h Data ta C Colle ollecte ted D d During EF3 uring EF3 Expe Experim riment nt Excelle Ex llent Oppor nt Opportunity to C tunity to Calibr librate te N Num umeric rical Mode l Models ls Ø Ele Electric trical Mode l Modeling ling : calibrate model to match electrical measurements • derive distribution of graphite connections on the • fracture predict heat input distribution on the fracture • Ø T The herm rmal Mode l Modeling ling : calibrate model to match measured temperatures • derive heat input distribution • estimate thermal diffusivity of surrounding rocks • 6

Ele Electric trical Mode l Model: EF3 l: EF3 Expe Experim riment nt • Develop a model to match electrical measurements, and use it to predict heat input distribution on the fracture Me Measur sured V d Volta oltages s .8 0.7 .7 0.0 .0 0.8 0.0 .054 0.9 .9 -0.4 -0 .4 0.6 .679 -0.8 -0 .8 -0 -0.0 .033 OB16 OB OB OB13 OB3 OB 1.0 .000 0.0 .027 0.9 .902 -0.7 -0 .773 -1 -1.0 .000 EF3 EF3 OB OB4 OB OB15 OB9 OB OB OB5 OB2 OB OB OB17 OB OB14 0.6 .677 0.9 .904 0.3 .330 • Distribution (thickness) of calcined coke in the fracture from core Cok oke f fille illed fr d fractur ture -0.796 -1.000 7

Ele Electric trical Mode l Model: EF3 l: EF3 Expe Experim riment nt Spatial distribution of graphite thickness tuned to • match observed voltages Me Measur sured V d Volta oltages s .8 0.7 .7 0.0 .0 0.8 0.0 .054 0.9 .9 -0 -0.4 .4 0.6 .679 -0.8 -0 .8 -0.0 -0 .033 OB OB16 OB13 OB OB3 OB 1.0 .000 0.0 .027 0.9 .902 -0 -0.7 .773 -1 -1.0 .000 EF3 EF3 OB OB4 OB15 OB OB9 OB OB5 OB OB2 OB OB OB17 OB OB14 0.6 .677 0.9 .904 0.3 .330 Mode Modele led V d Volta oltage D Distrib istribution ution -0.0 -0 .041 0.7 .729 -0 -0.0 .0432 1.0 .000 -0 -0.0 .038 0.9 .923 -0 -0.7 .796 -1 -1.0 .000 OB9 OB4 -0 -0.7 .742 0.8 .895 0.2 .282 • Excellent agreement between modeled and measured voltages 8

Ele Electric trical Mode l Model: EF3 l: EF3 Expe Experim riment nt A F Forw orward Mode d Model to Pr l to Predic dict H t Heat Input D t Input Distrib istribution ution • Heat input distribution predicted from distribution of conductive material and voltage Modele Mode led H d Heat Input D t Input Distrib istribution ution Btu/ft 2 -hr 30 20 OB9 10 OB4 0 • Predictions for different time slices suggest little temporal variation in heat input distribution ü Ele Electric trical Mode l Modeling c ling can be n be use used to pr d to predic dict he t heat input distrib t input distribution ution fr from om obse observed v d volta oltage distrib distribution on the ution on the fr fractur ture 9

The herm rmal Mode l Model: EF3 l: EF3 Expe Experim riment nt An In n Inverse se Mode Model to Pr l to Predic dict H t Heat Input D t Input Distrib istribution ution • Develop a thermal model to match measured temperatures at different times in the 90-day experiment Fibe iber Optic r Optics H s Hole oles s The herm rmoc ocouple ouple H Hole oles s -1.000 • Estimate thermal diffusivity of surrounding rocks and derive heat input distribution 10

The herm rmal Mode l Model: EF3 l: EF3 Expe Experim riment nt Deriv rived H d Heat Input D t Input Distrib istribution ution Btu/ft 2 -hr OB9 OB9 OB4 OB4 Perpendicular Parallel Good agreement between modeled and measured temperatures 11

The herm rmal Mode l Model: EF3 l: EF3 Expe Experim riment nt Predicted temperature after 90 days º F • Sensitivity analysis performed to estimate thermal diffusivity • 1-D model developed to estimate thermal diffusivity from temperature fall-off data • Estimated thermal diffusivity range, 0.45 - 0.6 ft 2 /day, agrees with lab and field measurements 12

Calibr librate ted Mode d Models: EF3 ls: EF3 Expe Experim riment nt Predicted Heat Input Distributions Btu/ft 2 -hr Electrical Model 30 20 10 OB9 0 OB4 Thermal Model 32 24 16 OB9 8 OB4 0 Heat input distributions predicted by electrical and thermal models have same general attributes 13

lusions: Conc onclusions: Calibr librate ted Mode d Models: EF3 ls: EF3 Expe Experim riment nt Field experiment verified that electrically conductive fracture Ø can be constructed in the field and operated at low temperatures Numerical models developed to analyze thermal and Ø electrical data collected during the experiment Electrical model calibrated to match voltage measurement Ø and to predict the heat input distribution on the fracture Thermal model calibrated to match measured temperatures Ø and to derive heat input distribution Heat input distributions predicted by the two models have Ø same general attributes 14

Ong Ongoing W oing Wor ork: Calibr librate ted Mode d Models: EF1 ls: EF1 Expe Experim riment nt Me Measur sured V d Volta oltages s 0.2 .25 -0.6 -0 .64 0.3 .34 -0.9 -0 .94 0.2 .28 0.8 .81 0.9 .99 1.0 .00 -0.9 -0 .90 Thermal modeling -1.0 -1 .00 0.7 .78 0.3 .33 -0.5 -0 .50 -0 -0.9 .90 0.6 .63 20 F Feet t 15

Acknowle nowledgm dgments nts ExxonMobil Oil Shale: Michele Thomas, William Meurer, Alex Morelos, Jesse Yeakel, Ana Carmo, Norman Pokutylowicz, and Matthew Spiecker 16