Here we report the key success factors of the first polymer pilot at YPF in the south of Argentina and the consensus shifting strategy for polymer expansion in the current oil price context. We calculated water flow velocities in the reservoir using three multiscale history matched simulation models. We found that more than 80% of water velocities across the complete field are below1ft/day [normally assumed reservoir water velocity for calculating the resistance factor in laboratory experiments].We increased polymer concentration in 10 to 30% to ensure good mobility ratio in the high permeability streaks possibly located in the channel bars. This is very important because low end point water permeability (<0.09) could explain the success of water flooding in friable formation with viscous oil. This challenges the common assumption of poor performance because adverse mobility ration (> > 10). High permeability streaks (above 10 Darcy) are not characterised because they are often lost during coring or they are not suitable for coreflooding experiments. Instead, stochastic history matching supports the idea of greater water end point permeability (>0.2) in the high perm streaks. Then, the target resistance factor for polymer could be underestimated (underestimation of polymer concentration) and polymer injection might not perform as expected. Simulation-based analysis of flows in the pilot zone strongly suggests that one of the key success factors was pattern confinement. The pilot configuration is five-spot with 4 injectors, 1 confined producer and 9 offsets producer wells. After injecting 0.15 pore volumes of 2500/3000ppm polymer we recovered more than 11% ooip incremental oil above waterflooding from the central pattern and more than 6% of the ooip from offset producers in contacted zone. The water cut reduced from 90% to 45% in the confined producer and from 87% to 67% in the offset producers. Water cut reduction and therefore the oil response is greater than best pilots in the literature. This can be explained because of the cross flow between fluvial layers in the inter-well region. Our simulations indicated that there was no out-flow of the central pattern. The very good performance in terms of low utility factor obtained so far (2.9 kg per incremental barrel of oil above water flooding) supports the hypothesis of the good confinement. The accurate simulation model allowed us to conceptualise a pattern rolling strategy for the polymer expansion that makes this technology economic for this low oil price context.

After the successful pilot (18%STOOIP incremental oil, Juri et al. 2017 and utility factor of 1,76 kg polymer/bbl) of two technologies, continuous polymer injection 0.3PV followed by high frequency water-alternate-polymer 0.3PV, a series of multiple simulations cases indicated an optimal extension of 3-year cycle -0.4 to 0.6PV- factory mode development. Despite all advantages of polymer injection, few full-field expansions emerge from many challenges in integrating multiple disciplines -lab studies, subsurface reservoir modelling, engineering, energy, procurement, logistics, storage, project management- all parties find it difficult focusing on a common goal and streamline the decision process. Because it has long been thought that the EOR road map means planning for large scale EOR. However, that is false because there are very few full-field injections. The past strategy in EOR has led to a few pilot extensions implemented. Planning for large-scale EOR suffers from having the incremental oil, uncertainty in oil price, uncertainty in economic drivers and uncertainty in technical solutions for the expansion on the same time frame, which undermines economics. What it seems to be the only way to make progress with EOR, in reality, is that EOR should align with planning for quick oil response. We present here the steps to accelerate EOR in Grimbeek field from a 4-injectors pilot to 80 new injectors in a fast, less than 18 months deployment. We compare the capital and operating cost of multiple expansion scenarios. The scope is to accelerate EOR deployment opposite to the slow typical EOR workflow. We have seen the opportunity in accelerating the oil focusing on a simple modular strategy rather than iterating multiple possible engineering solutions to distribute the polymer across the field. NPV of the overall project is higher than the costs of delaying the injection or having to redo parts of the implementation along the deployment. Future optimisation is necessary for options of logistics, storage, in-situ skid connection and mounting along with by-pass remote control valves and remote control of skid injections. The building blocks of the strategy are containerised mobile skids containing ten pumps, polymer dispersion unit, and 30-metric ton silos. The mobile skids plug-in the waterflooding manifolds in an easy and fast connection. This injection scheme distributes polymer using the already implemented waterflooding in-situ installation. We focus on having a simple, standard and mobile system targeting fast implementation and fast oil response rather than large facilities.

No abstract

No abstract

The aim of this study was to reduce risk in polymer flooding projects. Specifically, this study investigates polymer retention and viscosity losses. The study reservoir is a fluvial type reservoir with friable/unconsolidated sandstone units that are saturated with viscous oil. The reservoir is located in the south of Argentina. Retention and viscosity losses are the key parameters in designing a polymer flooding. They slow down the polymer velocity and deplete the polymer slug leading to a significant impact in the oil banking formation and therefore in the oil response. In friable/unconsolidated reservoir rock samples the usual laboratory procedure for retention estimation is fraught with complications and errors and may lead to unreliable results. We used a 20 month single-well injection/back-production test. And we monitored the sweep efficiency with time-lapse formation resistivity logging. The well schedule included, 5 months of water injection, three months of polymer injection and 12 months of back-production. We used an innovative inline viscometer to measure viscosity in the backflow stream at anaerobic conditions. This procedure avoided the problems of chemical degradation due Fenton chemistry with the oxygen scavenger. We measured the concentration with two methods, COD (Chemical Oxygen Demand) and bleach method. The back-produced polymer solution showed relatively low viscosity losses as compared to that of the injected solution. The interpretation of the polymer concentration measurements with a detailed simulation model led to retention bounds between 35 to 260 micro-grams of polymer per gram of rock with heterogeneous retention depending on the facies. This is sevenfold the value estimated from packs of crushed reservoir core material and threefold the value from coreflood studies in reservoir samples. This study has allowed us to adjust the forward planning and understand the retention behaviors in distributed clay fluvial formations. This methodology can provide low cost reliable estimations of polymer retention at a larger scale than corefloods.