6781097272

The reservoir is a weakly Consolidated sandstonc reservoir. faultcd and slightly dipping towards thc southcast with a top depth ranging between 560-885 mcters. The lliickncss of the productive Bentheimer sandstonc layers ranges between 20-35 meters, consisting of two layers, 10-20 meter gross each, separated by a shale layer. The sandstonc lias a porosiły of 20-30% and pcrmeability front 100-5000 mD. Initial rcscrv'oir prcssure at 750 m was 80 bar. corrcsponding to a prcssurc gradient of 0.11 bar/nt. Currently about 25% of the STOI1P of 104 million ton has bccn rccovcrcd. reducing thc actual prcssure gradient to 0.06-0.07 with 0.11 bar/tn close to thc aquifer.

The wells in tliis area produce between 5-20 tn3/d. For many wells in this area production is reduced because of flnes migration. In addition. thc oil viscosifies due to degassing of the oil, cspccially around thc wcllbore. Most of the wells are complctcd with wirc-wrapped sandscrccns, reducing thc sand/fines production to less tlian 0.5 g/1. The standard stimulation treatment that is being applied for thcsc wells is to inject hot water with a smali conccntration of surfactants and demulsiflcrs.

PATTERN SUMMARY AND ANALYSIS

The pattem choscn for the injcction pilot project is centred on the injector well 678. The six offset producers are well 86, 311. 677, 679. 688 and 689. The pattem configuration is shown below in Figurę 4. Using a field map of the field, the pattem area was calculatcd to be around 95,900 nf. Tablc 1 shows the resen oir parameters for the pattem wells.

The thickness of the different layers was obtained for the pattem in order to estimate a reprcsentative average value, to conduct Yolumetric calculations. and to enable creation of simplc geological cross scctions to visualize potcntial water flow paths from thc injector to thc producers. This data is shown in Tablc 2.

With an avcragc injcction ratę of 108 m³/d in well 678, and fluid off-takc of roughly 173 m 7d in the producers. the pattem voidagc replacement is approximatcly 62%. Offtake data for the pattem is shown in Table 3. Pattem production has been allocatcd 100% to the well 678 pattem due to the fact that other injectors are a minimum distancc of 500 m away and has at least one row of producers between thc well 678 pattem production wells aid thc next injector. The areał pattem production confonnance is presented in Figurę 5.

Using an averagc aggregate rcservoir thickness of 32.1 metres (upper and lowcr oil sand) for thc pattern area, a connatc water saturation of 20%, an irrcducible oil saturation of 25% and 100% areał and vertical sweep cfficicncy. the flooding ratę for tlie pattem can be calculatcd. Using these parameters a flooding ratę of 4.3% TPV/year (TPV - Total Porę Volume) or 5.3% HCPV/ycar (HCPV - Hydro Caibon Porę Volumc) was calculatcd. Another way of looking at the data is to see how long it takes to inject one porc volumc into the pattem. which cornes out as 23.4 years for one TPV and 18.7 years for one HCPV.

Figurę 6 sltows the front velocity and time for injected water to reach a given distancc from the injector assuming 100% areał and yertical sweep efficiency.

PULSE STIMULATION

Tliis scction has bccn diridcd into thrcc parts in order to discuss performance against each of the thrcc elements of the original sccpe of the project.

Incrcasing Injcctivity

To simplify analysis of pre and post PPT stimulation results, it was dccidcd to keep thc water injcction into well 678 constant, liinitcd to a maximum of 108 m7d during thc pilot. The cffcct of constraining the water supply is discusscd in rnore dctail in thc scction below on production.

Water injcction prcssure was stable just tindcr 30 barg prior to the PPT pilot project. Three weeks after the conlinuous pulse stimulation nad started, the wellhead injcction pressures had dropped to 11 barg. Over time, the injcction prcssure inereased somewhat. but stayed well below the pre-stimulation injection prcssure. The injcction prcssure following thc trial stayed at around 20 barg or around 28% lower tlian the steady stale injcction prcssure for around a month before slowly starting to risc againAs the injection ratę was held constant. thc injectivity indcx was calculatcd in order to craluatc thc injectivity enhancement during the pulse stimulation. Figurę 7 show s the water injection data.

Initially injectivity inereased by 40%. This is a significant inerease and corresponds to the lower injection pressures observcd. The injectivity indcx then dcclincd. possibly associatcd with a pressurization of thc ncar wcllbore region. The dcclinc in injccth ity also corresponds to an obseryed inerease in amiulus prcssure, indicating the beginning of a tubing leak. The corrosivc naturę of thc injected fluid and its effect on the tubing and PPT equipment were morę severe tlian anticipatcd which causcd the cquipmcnt to be heavily corrodcd at the end of the trial. The average injectivity enhancement for the trial was around 30% with maximum and minimum obseryed values of 40% and 15%. rcspectively.

Dcmonstrating tliis injectivity enlianccmcnt is scen as a very iinportant indicator for prcssure pulse stimulation, in that for a liquid fillcd system, oil production is dircctly proportional to water injection rates.

Effect on Production

The improvement in injectiyity could be related to an improyement in injection conformance and hence could translatc into inereased production and ultimatc oil rccoYcry for a conlinuous PPT installation.

The production data from each of thc wells in thc pattem is