No abstract

A crosswell electromagnetic (EM) tomography survey was acquired between two horizontal wells, a water injector and a producer, using coil tubing conveyance. The survey was conveyed in a swept area of a naturally fractured Middle Eastern carbonate reservoir. This paper presents the work flow that was developed to integrate the EM survey results with other data to successfully map interwell hydrocarbon saturations. A slanted well was later landed in the interwell area based on the crosswell survey interpretation results. This new well targeted an oil column shown to be bypassed by the water injection in the swept area, and the well is currently on production. The survey established a number of world firsts for the EM technology, including horizontal wells deployment, well spacing >1.3 km, full 3D resistivity inversion and 3D apparent water saturation mapping. The field program consisted of an extensive openhole logging sequence in each well prior to the crosswell EM data collection. The crosswell EM survey measured 135 profiles, covering more than an 800 m horizontal section in each of the two wells. High quality data were acquired for almost all the data profiles, even in a 400 m cased section. The field data were interpreted with a 3D inversion code. The fracture corridors were identified as electrical conductive volumes, due to their high water saturation. The less or unswept areas were characterized by higher resistivity, due to their lower water content. A resistivity model cannot be directly used for reservoir evaluation, then the subsequent step was to convert the resistivity model into a saturation volume, based on Archie's law. An innovative workflow of petrophysical simulations was designed to define volumetric distribution of porosity and water salinity based on reservoir simulation scenarios and ad hoc laboratory analysis. Finally, the most probable scenario of apparent water saturation volume in the interwell reservoir volume was produced. This work outlines the critical need for deep characterization in cases of fractured carbonate reservoirs, and in any case where the sweep efficiency is likely to be uneven, due to heterogeneities of reservoir petrophysical properties, and particularly permeability field variations.

Forecasting the estimated ultimate recovery (EUR) for extremely tight gas sites with long-term transient behaviors is not an easy task. Because older, more established methods used to predict wells with these characteristics have shown important limitations, researchers have relied on new techniques, like long short-term memory (LSTM) deep learning methods. This study assesses the performance of LSTM estimations, compared to that of a physics-based reservoir simulation process. With the goal of obtaining reliable EUR forecasts, unconventional tight gas reservoir data is generated via simulation and analyzed with LSTM deep learning techniques, tailored for sequential data. Simultaneously, a reservoir simulation model that is based on the same data is generated for comparison purposes. The LSTM forecasting model has the added benefit of considering operational interventions in the well, so that the machine learning (ML) framework is not disrupted by interferences that do not reflect the actual physics of the production mechanism on well behavior. The comparison of the data-driven LSTM deep learning model and the physics-based reservoir simulation model estimations was performed using the latter framework as a benchmark. Findings show that the AI-assisted LSTM model provides predictions similarly accurate to the ones estimated by the physics-based reservoir model, but with the added capability for long-term forecasting. These data-driven EUR models show great promise when analyzing unusually tight gas reservoirs that feature time series well information, which can improve estimations about recovery and point engineers towards better decisions regarding the future of reservoirs. Therefore, exploring deep learning methods featuring varying types of artificial neural networks in greater detail has the potential to significantly benefit the oil and gas sector. When compared to other machine learning methods, novel deep learning techniques have advantages that remain underexplored in the literature. This paper helps fill this gap by providing a valuable comparison between older prediction methods and new estimation simulations based on neural networks that can predict long-term behaviors.

Accumulation of gas-condensates in near wellbore regions, and consequently reduction of gas production rates, is a well-known phenomenon in most gas reservoirs all around the globe. In this study, injection of CO2 in "huff-n-puff" mode and its potential advantages of removing accumulated gas-condensates are thoroughly investigated. A compositional simulator was used to capture the near wellbore fluid flow physics causing the condensate dropout under various wellbore/field operational conditions. The severity of condensate blockage was evaluated using detailed sensitivity and uncertainty analyses of various key static and dynamic influencing parameters such as porosity, permeability, fluid composition, boundary flux, and flowing bottomhole pressure. Under various scenarios of condensate blockage, studies related to effectiveness and optimization of CO2 slug-size, soaking-time, and flowback rate are conducted. As prove of concept, it is observed that miscibility of CO2 in condensates, and near wellbore re-vaporization of accumulated liquids, are two of the main mechanisms of removing the condensate blockage. It is also concluded that CO2 is relatively more effective in removing near wellbore condensates as compared to far-field regions. Finally, based on various optimization analyses, a viable approach to adopt huff-n-puff CO2 injection scheme is suggested.

No abstract