As the use of hydrogen gas (H2) as a renewable energy carrier experiences a major boost, one of the key challenges for a constant supply is safe and cost-efficient storage of surplus H2 to bridge periods with low energy demand. For this purpose, underground gas storage (UGS) in salt caverns, deep aquifers and depleted oil-/gas reservoirs has been proposed, which provide suitable environments with potentially high microbial abundance and activity. Subsurface microorganisms can use H2 in their metabolism and thus may lead to a variety of undesired side effects such as H2 loss into formation, H2S build up, methane formation, acid formation, clogging and corrosion. We present a new AI framework for the determination of metabolism processes of subsurface microorganisms in hydrogen underground storage. The AI framework enables to determine the potential microbial related processes and reactions in order to optimize storage strategies as well as incorporate potential remediating actions to ensure efficient and safe underground hydrogen storage and minimize related side effects. We evaluated the framework on investigating potential microbial reactions for hydrogen storage in the Pohokura gas field in New Zealand. The framework evaluates reservoir parameters, such as temperature, pressure, salinity and hydrogen injection volumes as well as duration, and then classifies which reactions may take place as well as indicates the likelihood of the reaction taking place. For the deep learning framework, an optimized random forest algorithm was implemented and compared to other multi-class classification problems. The results demonstrated the ability to determine the microbial reactions that may take place with hydrogen storage reservoir as well as its severity, which enhances the optimization of injection strategy as well as suitability of a reservoir. This framework represents an innovative approach to microbial reaction prediction for underground hydrogen storage. The framework allows potential reservoirs to be efficiently evaluated and optimization strategies to be utilized in order to maximize the efficiency of underground hydrogen storage.
CO2 has some critical technical and economic reasons for its use as an injection gas for oil recovery. CO2 is very soluble in crude oil at reservoir pressures; it contributes to sweep efficiency enhancement as it swells the oil and significantly reduces its viscosity. Although the mechanism of CO2 flooding is the same as that for other gases, CO2 is easier to handle, it is cheaper, and it is an environmentally better candidate than other gases. Formation evaluation and reservoir engineering have been major areas in the oil and gas industry that are heavily influenced by technology advances, to increase efficiency, improve hydrocarbon recovery and allow real-time reservoir monitoring. Water flooding for increasing oil recovery has been amongst the oldest production mechanisms widely utilized since the end of the 19th century to maintain pressure levels in the reservoir and push hydrocarbons accumulations towards the producing wellbore locations (Satter, Iqbal, & Buchwalter, 2008). Produced water from the reservoir formation was reinjected in order to maintain pressure levels, as well as seawater and aquifer water injection have also taken a strong mandate. With the advent of technology and processing plants this injection process was further refined, allowing salinity control of the injected water as well as monitor the injection and distribution of the water levels in near real time (Boussa, Bencherif, Hamza, & Khodja, 2005). Formation evaluation has seen an even greater penetration of technology in its area with the quest to achieve real-time formation evaluation during the drilling process. Conventional formation evaluation is conducted utilizing wireline logging technology, which is deployed after the drilling of the well and allows to analyze the reservoir formation. Given the significant advancement of logging technologies, acquiring the measurements during the drilling process (LWD) has been at the forefront of interest, allowing improved well placement and geosteering as well as real-time formation evaluation to optimize well completion strategies (Hill, 2017). Amongst the technologies recently deployed, surfaced logging and advanced mud and logging allow to determine on cuttings in real time mostly any of the properties previously possible only on direct measurements on cores (Santarelli, Marsala, Brignoli, Rossi, & Bona, 1998; Katterbauer & Marsala, A Novel Sparsity Deploying Reinforcement Deep Learning Algorithm for Saturation Mapping of Oil and Gas Reservoirs, 2021; Katterbauer, Marsala, Schoepf, & Donzier, 2021). With advances in AI, reservoir characterization is now moving towards real-time or near real-time analysis at the rig site. For near real-time analysis, the main physical source of data is drill cuttings as it guides the drilling operation by determining important depth point such as formation tops, coring intervals. Traditionally, the description of these cuttings is done manually by geologists at the well site. The accuracy of these descriptions can be variable depending on the geologist's experience and indeed their mental state and tiredness level. Cores is another source of data. New techniques and older techniques imbued with AI components new allow for greater automation, efficiency, and consistency. The use of AI on traditional images are of great interest in the oil and gas community as they are: 1) fast to acquire, and 2) do not typically require expensive hardware. For example, Arnesen and Wade used convolutional neural networks; specifically, an inception-v3 inspired architecture, to predict lithological variations in cuttings (Arnesen & Wade, 2018). In their study, each sample is related to one lithology. Buscombe used a customized convolutional neural network to predict the granulometry of sediments, specifically the grain size distribution (Buscombe, 2019). Similarly, automated core description systems (e.g., (Kanagandran; de Lima, Bonar, Coronado, Marfurt, & Nicholson, 2019; de Lima, Marfurt, Coronado, & Bonar, 2019) and microfossil identification systems (e.g., (de Lima, Bonar, Coronado, Marfurt, & Nicholson, 2019)) are also being explored using neural networks with varying degree of success. A comprehensive review on the state of usage of rock images for reservoir characterization presented by de Lima et al. (de Lima, Marfurt, Coronado, & Bonar, 2019). In addition, the community is also recognizing the potential of improving older techniques by integrating artificial intelligence into their workflow. In reservoir characterization, chemostratigraphic analysis X-ray fluorescence is a prime example for this especially with the difficulties encountered when analyzing mudrocks in shale plays using traditional methods. The rise of XRF measurement was also fueled by the introduction of highly portable XRF devices that take 10s of seconds to measure one sample. The use of artificial intelligence techniques is being studied. For example, fully connected neural networks are applied on XRF data to predict total organic carbon (Lawal, Mahmoud, Alade, & Abdulraheem, 2019; Alnahwi & Loucks, 2019). In addition to the traditional elemental to mineralogical inversion methods such as constrained optimization, neural networks are being utilized (Alnahwi & Loucks, 2019). The integration between XRF, X-ray diffraction (XRD) measurements (Marsala, Loermans, Shen, Scheibe, & Zereik, 2012), and well logs using traditional statistical methods and neural network methods is also being explored (Al Ibrahim, Mukerji, & Hosford Scheirer, 2019). The integration between artificial intelligence systems and automated robotic scanning systems (e.g., (Croudace, Rindby, & Rothwell, 2006)) is key in introducing these technologies into the daily rig operations. The low density of CO2 relative to the reservoir fluid (oil and water) results in gravity override whereby the injected CO2 gravitates towards the top of the reservoir, leaving the bulk of the reservoir uncontacted. This may lead to poor sweep efficiency and poor oil recovery; this criticality can be minimized by alternating CO2 injection with water or similar chase fluids. This process is known as Water Alternating Gas (WAG). A major challenge in the optimization of the WAG process is to determine the cycle periods and the injection levels to optimize recovery and production ranges. In this work we present a data-driven approach to optimizing the WAG process for CO2 Enhanced Oil Recovery (EOR). The framework integrates a deep learning technique for estimating the producer wells’ output levels from the injection parameters set at the injector wells. The deep learning technique is incorporated into a stochastic nonlinear optimization framework for optimizing the overall oil production over various WAG cycle patterns and injection levels. The framework was examined on a realistic synthetic field test case with several producer and injection wells. The results were promising, allowing to efficiently optimize various injection scenarios. The results outline a process to optimize CO2-EOR from the reservoir formation via the utilization of CO2 as compared to sole water injection. The novel framework presents a data-driven approach to the WAG injection cycle optimization for CO2-EOR. The framework can be easily implemented and assists in the pre-selection of various injection scenarios to validate their impact with a full feature reservoir simulation. A similar process may be tailored for other Improved Oil Recovery (IOR) mechanisms.
CO2 plume geothermal technology (CPG) has been developed in recent years by several companies. The technology aims to utilize CO2 stored in saline aquifers to produce geothermal energy. CPG is different from conventional geothermal concepts. Here, the feedstock utilizes CO2 as a carrier fluid through which heat is extracted from the subsurface reservoir. Furthermore, the system does not necessarily rely on shallow natural hydrothermal locations but can utilize a conventional sedimentary basis. At last, CPG can still harvest energy in low-temperature environments that are currently not suitable for conventional geothermal extraction. We present a new deep learning optimization framework for the maximization of power generation from a CPG system. The framework utilizes an adapted N-BEATS approach. The approach is based on a stack of ensembled feedforward networks that are also stacked by interconnecting backcast and forecast links. The advantages of the framework are its flexibility with respect to different input parameters and various forecastable time series. This is particularly important for CPG to easily capture variations in the temporal dynamics and temperature responses across the various CO2 injection and production wells. We evaluated the framework on a simulated CO2 storage reservoir based in the Taranaki basin in New Zealand. The Taranaki basin has been well studied for CO2 storage, given the presence of a large saline aquifer that may be well suitable for both CO2 storage and CPG energy production. We simulated 3.5 years of CO2 injection and production for geothermal energy production as input to the N-BEATS framework. The training performance of the network was strong, and the model's performance was then evaluated on subsequent two years of energy production. The deep learning framework is then integrated into a global optimization framework to optimize energy production while adapting CO2 injection. The new deep learning N-BEATS optimization framework for CPG power generation represents an innovative way to enhance energy generation from CO2 storage reservoirs providing a sustainable way to minimize carbon footprint while delivering energy.
Hydrogen has become a very promising green energy source that can be easily stored and transported, and it has the potential to be utilized in a variety of applications. Hydrogen, as a power source, has the benefits of being easily transportable and stored over long periods of times, and does not lead to any carbon emissions related to the utilization of the power source. Thermal EOR methods are among the most commonly used recovery methods. They involve the introduction of thermal energy or heat into the reservoir to raise the temperature of the oil and reduce its viscosity. The heat makes the oil mobile and assists in moving it towards the producer wells. The heat can be added externally by injecting a hot fluid such as steam or hot water into the formations, or it can be generated internally through in-situ combustion by burning the oil in depleted gas or waterflooded reservoirs using air or oxygen. This method is an attractive alternative to produce cost-efficiently significant amounts of hydrogen from these depleted or waterflooded reservoirs. A major challenge is to optimize injection of air/oxygen to maximize hydrogen production via ensuring that the in-situ combustion sufficiently supports the breakdown of water into hydrogen molecules. In-situ combustion or fireflood is a method consisting of volumes of air or oxygen injected into a well and ignited. A burning zone is propagated through the reservoir from the injection well to the producing wells. The in-situ combustion creates a bank of steam, gas from the combustion process, and evaporated hydrocarbons that drive the reservoir oil into the producing wells. There are three types of in-situ combustion processes: dry forward, dry reverse and wet forward combustion. In a dry forward process only air is injected and the combustion front moves from the injector to the producer. The wet forward injection is the same process where air and water are injected either simultaneously or alternating. Artificial intelligence (AI) practices have allowed to significantly improve optimization of reservoir production, based on observations in the near wellbore reservoir layers. This work utilizes a data-driven physics-inspired AI model for the optimization of hydrogen recovery via the injection of oxygen, where the injection and production parameters are optimized, minimizing oxygen injection while maximizing hydrogen production and recovery. Multiple physical and data-driven models and their parameters are optimized based on observations with the objective to determine the best sustainable combination. The framework was examined on a synthetic reservoir model with multiple injector and producing wells. Historical injection and production were available for a time period of three years for various oxygen injection and hydrogen production levels. Various time-series deep learning network models were investigated, with random forest time series models incorporating a modified mass balance – reaction kinetics model for in-situ combustion performing most effectively. A robust global optimization approach, based on an artificial intelligence genetic optimization, allows for simultaneously optimization of an injection pattern and uncertainty quantification. Results indicate potential for significant reduction in required oxygen injection volumes, while maximizing hydrogen recovery. This work represents a first and innovative approach to enhance hydrogen recovery from waterflooded reservoirs via oxygen injection. The data-driven physics inspired AI genetic optimization framework allows to optimize oxygen injection while maximizing hydrogen production.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.