Features of gas inflow to a horizontal wellbore at various trajectories

This article discusses the effect of wellbore trajectory on the flow performance of a horizontal cased and perforated gas well. We used a coupled well-reservoir flow model, taking into account the nature of the flow, and local hydraulic resistances of the wellbore, and thus determined the pressure and mass flow distribution along the horizontal wellbore for several types of trajectories, including undulated and toe-up trajectories. The simulation results showed the effect of horizontal gas well trajectory type on its flow rate and the importance of considering pressure distribution to optimize well design.

4D Seismic Detectability in a Brazilian Pre-Salt Carbonate Field

This paper is dedicated to a pre-salt carbonate field located within the Santos Basin, Brazil, comprising thick Aptian reservoirs interspersed with igneous rocks. One of the main challenges for reservoir management is the surface constraint on the gas, as all of the produced gas will have to be reinjected and can be miscible with the in-situ hydrocarbons. The recovery mechanism selected is mainly WAG (water alternating gas) injection, with both producers and injectors equipped with intelligent completions using Inflow Control Valves (ICVs). A 4D seismic monitoring survey is planned to delineate gas and water fronts in reservoir flow units about 10m thick, providing critical information to help piloting a planned 6-month WAG cycle for improved recovery. Seismic imaging is challenging in this case and 4D signal is expected to be weak (±2% dIp/Ip). We propose here, a methodology, based on a 1-D Gassmann fluid substitution model at wells (only limited reservoir fluid PVT data available) to rapidly answer the following pertinent questions as posed by the asset team in charge of the field: From a phenomenological stand-point and neglecting some possible processing, imaging and acquisition challenges, will 4D data (post 4D inversion) detect a gas streak from an injector to a producer? What is the 4D seismic detection limit based on reservoir thickness? What kind of seismic acquisition will assure this detectability? Under the assumptions made in this work, this methodology shows that a permanent system of acquisition seems to be a fit-for-purpose technology for detectability. Further work is however recommended using full complement of a 3D static and dynamic simulation model coupled with a complete fluid PVT model in order to assess more complex 3D dynamic interactions between the injectors and producers.

New method for Handling of Infrastructural Constraints for Integrated Modeling in Steady Case

Recently, more and more reservoir flow models are being extended to integrated ones to consider the influence of the surface network on the field development. A serious numerical problem is the handling of constraints in the form of inequalities. It is especially difficult in combination with optimization and automatic control of well and surface equipment. Traditional numerical methods solve the problem iteratively, choosing the operation modes for network elements. Sometimes solution may violate constraints or not be an optimal. The paper proposes a new flexible and relatively efficient method that allows to reliably handle constraints. The idea is to work with entire set of all possible operation modes according to constraints and control capabilities. Let's call this set an operation modes domain (OMD). The problem is solved in two stages. On the first stage (direct course) the OMD are calculated for all network elements from wells to terminal. Constraints are handled by narrowing the OMD. On the second stage (backward course) the optimal solution is chosen from OMD.

Simulating Multiphase Flow in Reservoirs with Generative Deep Learning

This study investigated the ability to produce accurate multiphase flow profiles simulating the response of producing reservoirs, using Generative Deep Learning (GDL) methods. Historical production data from numerical simulators were used to train a GDL model that was then used to predict the output of new wells in unseen locations. This work describes a procedure in which data analysis techniques are used to gain insight into reservoir flow behavior at a field level based on existing historical data. The procedure includes clustering, dimensionality reduction, correlation, in addition to novel interpretation methodologies that synthesize the results from reservoir simulation output, characterizing flow conditions. The insight was then used to build and train a GDL algorithm that reproduces the multiphase reservoir behavior for unseen operational conditions with high accuracy. The trained algorithm can be used to further generate new predictions of the reservoir response under operational conditions for which we do not have previous examples in the training data set. We found that the GDL algorithm can be used as a robust multiphase flow simulator. In addition, we showed that the physics of flow can be captured and manipulated in the GDL latent space after training to reproduce different physical effects that did not exist in the original training data set. Applying the methodology to the problem of determining multiphase production rate from new producing wells in undrilled locations showed positive results. The methodology was tested successfully in predicting multiphase production under different scenarios including multiwell channelized and heterogeneous reservoirs. Comparison with other shallow supervised algorithms demonstrated improvements realized by the proposed methodology, compared to existing methods. The study developed a novel methodology to interpret both data and GDL algorithms, geared towards improving reservoir management. The method was able to predict the performance of new wells in previously undrilled locations without using a reservoir simulator.