POSEIDON™ ROCM

POSEIDON™ ROCM  is a robust algorithm that allows to generate remaining oil maps honouring both the ‘known’ estimated localized phase distributions around the wells together with the material balance for each reservoir unit. It is naturally expected that this approach becomes increasingly reliable as reservoir maturity increases, and with greater production constraints (number of wells), but unlike reservoir simulation history-matching, the process remains very practical, resource-light and time-effective irrespectively of this increasing complexity.

rocm_1

POSEIDON ROCM Engine algorithm – a timestep overview

Reliable Remaining Oil Maps Without Simulation

The compliant mapping process consists of an integration of the following parts:

  • A material balance engine that ensures at the flow unit level that conservation of mass is computed, over the entire map. With pressure history and in-place volumes as given, aquifer influx time series are computed.
  • A mapping algorithm which ensures a realistic saturation or contact movement of water and gas, by integrating a 2D flow velocity solver with a gradual deformation algorithm to account for assumed geological trends and fractional flow values at each producer. In other words, the mapping algorithm is a proxy simulator which tries to mimic the water and gas encroachment into producing reservoirs
  • A coupled search algorithm which allows the computation of a ‘best’ solution that honours both material balance and the fractional flow behaviour of the wells at each time-step.
rocm_2

Remaining Oil Compliant Mapping. A significant step up from traditional analytical methods. Locating and quantifying remaining oil pools without full physics simulation.

Saturation Mapping

POSEIDON™ ROCM is able to perform the ROCM process with 2 levels of complexity:

  1. The 2D-only approach: in this configuration, the vertical heterogeneity is upscaled into a single set of averages (porosity, permeability, net-to-gross, net thickness, depth) and saturations are distributed accordingly
  2. The pseudo-3D approach (Advanced Fractional Flow): in this configuration, each well fractional flow is modified by the user-input log-data (Swi, Sgi, permeability vs depth) and the vertical heterogeneity in each cell is taken into account to distribute contacts throughout the reservoir, using the C-Track algorithm. When this option is selected a GOC and OWC contact in each well is calculated, and if relevant data is available, in each grid-cell of the map.
rocm_5

Advanced fractional flow modelling – capturing local well heterogeneity and monitoring contacts movement

Workflow Validation

The compliant mapping algorithm has been extensively validated with synthetic and real-life reservoir simulation models, to ensure its all-round robustness versus different geological and reservoir dynamics settings. The results show a consistent and remarkable fidelity versus the full physics 3D simulation.

rocm_6

Fluid contact (left) & STOIP maps (right) generated by ROCM (left) vs commercial simulator (right), for a Middle East major carbonate field case with 30+ horizontal wells.

Accuracy Vs Speed

When these two alternative methods are compared quantitatively for a number of randomly selected targets, the ROCM oil volumes distribution accuracy compared to simulation outcomes is consistently greater than 85% as shown in the figure below

rocm_4

Comparison of simulation results (grey) vs ROCM (blue) results for a number of target areas

Sensitivity & Risking

Thanks to its speed, the POSEIDON™ ROCM process can be rapidly iterated, thus generating a multi scenario set of outcomes. It is possible to investigate pragmatically the impact of allocation, geology and rock-fluid properties on the remaining oil distribution, and therefore to generate a risk profile for each identified LTRO opportunity. In particular, target risking can be performed using this process:

rocm_7

Uncertainty analysis capability with ROCM: generation of scenarios by combining rock-fluid property cases, roperty distribution cases and production allocation realizations

Synthetic Log Generation (C-track)

Estimated Gas Liquid Contact (GLC) and gas saturation using the C-Track production inversion algorithm. Note the very close match obtained to the simulation results (itself constrained to gridding size)

A production data inversed saturation log is generated to model vertical heterogeneity and assess in-well contacts versus time. A complex fractional flow model is also available.

If log data is available (porosity, permeability, Swi, Sgi) as well as facies-based relative permeability, for each of the producing wells, then the complex fractional flow evaluation option of PoseidonTM can be exploited to estimate the GOC (or GLC) and OWC at each producing well. Facies can be porosity or permeability bin-defined. The C-Track algorithm performs a production inversion which allows the accurate estimation of gas, oil and water distribution vertically within the well. This approach, illustrated on a synthetic simulation example below, allows the identification and evaluation of gas and water shutoff opportunities without direct in- ell measurements such as production logs (PLTs) or saturation logs (RST, PDK…). This approach can also be used to drive reservoir surveillance activities. The other benefit of deploying the C-Track algorithm is to model more accurately each well’s fractional flow, thereby taking into account the vertical heterogeneity distribution and its impact on GOR and WCT development as a function of volumetric invasion by gas and water.

Understanding contact movement assist work-over, intervention and surveillance planning and decision-making

  • Water/Gas shut-off: Gas or water invaded zones can be shut-off to reduce high water or gas production
  • Re-perf: Knowing current gas and water contact location allows re-perf in unwept zones
  • Stimulation: Acid or chemical can be injected in the intended zones
  • Data Acquisition: Certain data acquisition can be well planned with current contact information
rocm_3

Comparison between traditional method with ROCM mapping engine in POSEIDON