An artificial intelligence method for improving upscaling in complex reservoirs

Santos, Anderson Carlos Sousa e; Scanavini, H.F.A.; Pedrini, Hélio; Schiozer, Denis José; Munerato, Fernando Perin; Barreto, Carlos Eduardo

doi:10.1016/j.petrol.2021.110071

Cited by 9 publications

(4 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Lossy information compression is closely related to machine learning (note that renormalization group theory is closely related to deep learning (Mehta & Schwab, 2014)) and can be formulated using Bayesian inference (Cheng et al, 2018;Theodoridis, 2015). Additionally, machine learning has also been proposed for efficient upscaling (Santos et al, 2022), while Bayesian inference is frequently used to conduct model-data fusion and estimate model parameters based on observational constraints Here the subscale model represents processes in a spatial subset of the upscaled model. To indicate that models at scales l and l 1 may have different number of state variables and processes (due to lossy information compression from scale l 1 to scale l), subscripts n l and k l are used to indicate variables and processes at scale l, while subscripts m l-1 and j l-1 are used at scale l-1.…”

Section: Modeling Biogeochemical Processes Across Scalesmentioning

confidence: 99%

“…Lossy information compression is closely related to machine learning (note that renormalization group theory is closely related to deep learning (Mehta & Schwab, 2014)) and can be formulated using Bayesian inference (Cheng et al., 2018; Theodoridis, 2015). Additionally, machine learning has also been proposed for efficient upscaling (Santos et al., 2022), while Bayesian inference is frequently used to conduct model‐data fusion and estimate model parameters based on observational constraints (Tang & Zhuang, 2009; Vrugt, 2016). Therefore, in the following, the Bayesian framework (Jaynes, 2003) is used to explain the necessity and benefit of coherent scaling in formulating EBMs.…”

Section: Biogeochemical Processes In Terrestrial Ecosystem Dynamics A...mentioning

confidence: 99%

See 1 more Smart Citation

Feasibility of Formulating Ecosystem Biogeochemical Models From Established Physical Rules

Tang,

Riley,

Manzoni

et al. 2024

JGR Biogeosciences

View full text Add to dashboard Cite

To improve the predictive capability of ecosystem biogeochemical models (EBMs), we discuss the feasibility of formulating biogeochemical processes using physical rules that have underpinned the many successes in computational physics and chemistry. We argue that the currently popular empirically based approaches, such as multiplicative empirical response functions and the law of the minimum, will not lead to EBM formulations that can be continuously refined to incorporate improved mechanistic understanding and empirical observations of biogeochemical processes. Instead, we propose that EBM parameterizations, as a lossy data compression problem, can be better formulated using established physical rules widely used in computational physics and chemistry, and different biogeochemical processes can be more robustly integrated within a reactive‐transport framework. Through several examples, we demonstrate how mathematical representations derived from physical rules can improve understanding of relevant biogeochemical processes and enable more effective communication between modelers, observationalists, and experimentalists regarding essential questions, such as what measurements are needed to meaningfully inform models and how can models generate new process‐level hypotheses to test in empirical studies. Finally, while empirical models with more parameters are often less robust, physical rules‐based models can be more robust and show lower predictive equifinality, stemming from their enhanced consistency in representations of processes, interactions and spatial scaling.

show abstract

Section: Modeling Biogeochemical Processes Across Scalesmentioning

confidence: 99%

“…Lossy information compression is closely related to machine learning (note that renormalization group theory is closely related to deep learning (Mehta & Schwab, 2014)) and can be formulated using Bayesian inference (Cheng et al., 2018; Theodoridis, 2015). Additionally, machine learning has also been proposed for efficient upscaling (Santos et al., 2022), while Bayesian inference is frequently used to conduct model‐data fusion and estimate model parameters based on observational constraints (Tang & Zhuang, 2009; Vrugt, 2016). Therefore, in the following, the Bayesian framework (Jaynes, 2003) is used to explain the necessity and benefit of coherent scaling in formulating EBMs.…”

Section: Biogeochemical Processes In Terrestrial Ecosystem Dynamics A...mentioning

confidence: 99%

Feasibility of Formulating Ecosystem Biogeochemical Models From Established Physical Rules

Tang,

Riley,

Manzoni

et al. 2024

JGR Biogeosciences

View full text Add to dashboard Cite

show abstract

“…Mohaghegh identified successful practices in hydraulic fracturing by intelligent data mining tools and intelligence techniques and selected the stimulation of gas storage wells with neural network and genetic algorithm (GA) [45][46][47] . Anderson et al used an AI method for improving the upscaling of the geological model of complex reservoirs [48] . ANN, ANFIS, GA, GA-ANN, particle swarm optimization (PSO) have been utilized to predict the rock mechanical parameters, such as brittleness index, compressive strength, elasticity modulus and Poisson's ratio [49][50][51] .…”

Section: Research Progress Of Hydraulic Fracturing Numerical Simulationmentioning

confidence: 99%

A review on the application of cohesive zone model in hydraulic fracturing

Zhao

et al. 2023

IOP Conf. Ser.: Earth Environ. Sci.

View full text Add to dashboard Cite

Hydraulic fracturing is an effective measure to increase production and injection and blockage removal in oil and gas field development. Accurate prediction of fracture morphologies is the key to the optimized design of hydraulic fracturing. The cohesive zone model (CZM) has been widely used in the numerical simulation of fracture initiation and propagation during hydraulic fracturing. The fractures formed by the numerical simulation vary significantly with different CZMs. In the current numerical simulation, the CZM generally adopts the bilinear model, which is more suitable for describing brittle fracture, while rocks are quasi-brittle materials and have nonlinear CZMs. This deviation should be corrected. Moreover, the CZM parameters are generally determined based on experience, without a reliable basis and standard determination method. This article focused on the CZM, systematically introduced its concept and classification, and clarified the correlation between the types of CZMs and the brittleness, quasi-brittleness, and ductility of rock fracture. The application of CZM in hydraulic fracturing was reviewed, and the existing problems, corresponding countermeasures and future research trends were presented. An integrated method of combining laboratory experiments, data mining and numerical simulation to determine the CZMs of mode I, mode II, and I/II mixed mode cohesive cracks was proposed.

show abstract

“…The study of fluid flow often involves the analysis of carbonate rocks due to their properties (SANTOS et al, 2022). Pores and fractures, the main components of carbonate rocks, can provide important insights into reservoir exploration.…”

Section: Related Workmentioning

confidence: 99%

A Computational Tool for Geometric Characterization of Pores and Fractures in Microtomography Rock Images

Bomfim,

Avansi,

Vidal

et al. 2023

Rev. Bras. Cartogr.

View full text Add to dashboard Cite

Pores and fractures are important structures for the study and characterization of fluid flow inside rocks. It is through them that there can be the propagation and conduction of fluids and chemical substances inside the reservoirs, since the connections between these spaces provide the existence of flow. To understand the behavior of these structures, we have developed a computational tool that, through the use of high resolution images derived from computed microtomography, aims to detect and characterize the geometry of pores and faults with the objective of being an environment that allows the detailed understanding of these structures. For this task, image processing techniques were applied to identify the structure of the contours, obtained through data segmentation, and to evaluate their geometric parameters in conjunction with the algorithms implemented in this work. With this, we seek to provide a tool capable of assisting in the characterization of reservoirs and fluid analysis, and to reduce the uncertainties found in manual work through a computational approach, prioritizing the preservation of important samples collected in the field.

show abstract

An artificial intelligence method for improving upscaling in complex reservoirs

Cited by 9 publications

References 15 publications

Feasibility of Formulating Ecosystem Biogeochemical Models From Established Physical Rules

Feasibility of Formulating Ecosystem Biogeochemical Models From Established Physical Rules

A review on the application of cohesive zone model in hydraulic fracturing

A Computational Tool for Geometric Characterization of Pores and Fractures in Microtomography Rock Images

Contact Info

Product

Resources

About