Placing horizontal wells in the correct zones of the producing reservoir in static/dynamic models is important for robust model quality and essential for production history matching. A laborious technique of manually generating correction points around each un-calibrated well is often used by geomodelers. This technique is not systematic and is highly interpretive. We present an automated global solution that leverages not only the well tops, but also the well trajectory information to calibrate the horizontal wells. The solution comprises a series of processes that can all be run through a custom built graphical user interface (GUI). The processes are designed to a) detect the calibration problems with the ability to visualize them, b) correct each surface separately for zone mismatch and c) retrieve original zone thickness whenever possible. Treating each surface separately simplifies the problem and causes great reduction in run time compared to simultaneous correction of all surfaces in one go. However, the final results are essentially global, which is insured by maintaining the input zone thicknesses (except if well data confirms otherwise). The correction is achieved by ensuring that model zones (between the input surfaces) match the "optimum zone logs" at all wells in the model. The term optimum, in this context, means that the zone log should only contain the zones to be modeled, no more no less. These optimum zone logs are typically calculated from the well tops, which implies having an impeccable set of well tops. In most cases, well tops sets may contain extra tops (that are not included in the model) and missing tops at some of the wells. While the latter problem is more difficult to solve, both problems must be detected and corrected for a successful run of this solution. Fortunately, the "diagnosis" step at the beginning of this solution detects all these problems, which reduces the time needed to review the well correlation, especially in fields with hundreds of wells. Moreover, machine learning techniques are used to suggest the position of the missing well tops. The user gets to either approve the suggestion or change it manually. The solution we are presenting is fully automatic and fully reproducible. It is given with four parameters to control the amount and influence of the correction applied to the horizon intersecting the trajectory when it is not supposed to. If optimum zone logs are ready, the solution can run in few minutes to correct large models with hundreds of wells and tens of surfaces. This has the potential of reducing months of work to few days to calibrate horizontal wells in a structural model. The few days may include optimization of the solution parameters to achieve the desired results.

No abstract

Thanks to high-resolution seismic enabled by hardware and software technology advances, 3D seismic surveying techniques have become particularly effective in hydrocarbon prospection and development. This paper, based on two seismic datasets, well data and production history from deep-offshore Angola, Block 4, highlights how seismic data may be used for both reservoir characterization (static model and in-place) and reservoir monitoring (saturation changes). In the following we shall present in detail the applied seismic characterization method, with particular emphasis on: (i) advanced data conditioning and modeling, (ii) sequential elastic inversion for the integration of two datasets of different vintages, (iii) utilization of Bayesian techniques for reservoir characterization and probability density functions and finally (iv) Neural Network application for the assessment of oil saturation. The high resolution seismic data contains strong AVO signature, but to exploit these fully, preserving the signal throughout the process is essential: (i) seismic inversion feasibility analysis and synthetic gathers confirmed the quality of the far offset up to 55 degrees; (ii) this allows producing reliable density inversion products, using data beyond 35 degrees angles with adequate angle bands (iii) to get best inversion results a tailored pre-stack data conditioning was applied using five partial stacks with particular attention to amplitude preserving and flattening at reservoir level; (iv) finally well and seismic results were integrated through inversion and litho-classification, which compared well with well data. The high-fidelity inversion results allowed trying a novel approach through an additional step: saturation estimation based on Neural Network (NN) technique: (i) a labeled data set of elastic well logs, namely, P-wave Impedance, P-wave and S-wave velocities ratio and Density were used to train a Neural Network engine to estimate the water saturation from petrophysics. A low level of cross-validation error was achieved and deemed acceptable. The trained NN model seemed to be able to estimate the Water Saturation logs accurately as confirmed by blind well tests. A saturation cube was generated by applying the trained NN model on the three properties established through pre-stack inversion. Owing to the excellent quality of the recent (2018) high resolution seismic data, and despite log quality issues (poor borehole condition at some wells), high-fidelity elastic inversion could be achieved for both datasets. This in-turn led to a significant reservoir characterization uplift (additional in-place, better segregation and distribution of reservoir facies for simulation). Finally, the comparison between inversion results from the older dataset (prior to production) and the recent one allowed highlighting saturation variations pointing to swept and unswept portions of the field.

The acquisition of high-resolution and high-fidelity seismic data opens the possibility of precision reservoir characterization, particularly for those formations that do display a very pronounced Amplitude Versus Offset (AVO) response. In a traditional approach, the integration of seismic and well data information is driven by the ideas and view of the geologist, whose ideas and views might introduce bias in reservoir representation. It is undeniable that the guidance of the model construction by the geologist is an absolute requirement where seismic data limitations (quality, fidelity or accuracy) does not allow extracting a high-resolution reservoir description as an input to the static model construction. This is evolving: Where AVO effects at reservoir level are significant, the grade of both seismic data and petrophysical measurements now available allow a very precise description of the reservoir, in terms of reservoir geometry and reservoir property distribution. In the study area, AVO response in the Upper Miocene turbiditic deposits, offshore Angola are suitable for detail reservoir characterization. In this paper, we shall demonstrate through a real case how the results of seismic inversion and attribute analysis have been used for the characterization of depositional elements and the modeling of reservoirs. We shall present first how seismic inversion results have been integrated with facies defined from logs, and how this was used to confirm and identify fluid type and contacts verified at wells, as well as for the delineation of structural and stratigraphic reservoir boundaries. A systematic approach to constructing the 3D static model was applied. The top reservoir was picked in detail, allowing for the channel geometries to be sculpted out using seismic inversion results. The integration of qualitatively and quantitatively delineated sand-body architecture, geometry and orientation from seismic data. Sequential Indicator Simulation method, using well data, allows the facies to be modeled hierarchically through a combination of deterministic and stochastic approaches. The advantages of this methodology are: (i) integrating fully facies characterization and geological concepts with seismic response, (ii) reducing uncertainty on reservoir distribution and extent and, (iii) constructing of a coherent reservoir description fit for dynamic model even with limited number of wells.