Hasil untuk "Petroleum refining. Petroleum products"

Jaber B. Al Jaberi, Badr Bageri, Salaheldin Elkatatny

Abstract In high-density water-based drilling fluids, optimizing filtration behavior, filter cake characteristics, and rheological properties is essential for efficient and safe drilling operations. This study investigates the impact of perlite addition on these critical properties using three mud formulations with varying densities. Filtration tests were conducted using filter paper and ceramic disks, while rheological properties were evaluated using industry-standard models. Results revealed that increasing mud density leads to higher filtration volumes and thicker filter cakes, which potentially compromises wellbore stability and increasing the risk of pipe sticking. However, the incorporation of perlite significantly enhanced performance by reducing both filtration volume and filter cake thickness across all densities. Rheological models’ assessments illustrated that the perlite had minimal impact on plastic viscosity (PV) and yield point (YP), but a notable improvement in gel strength behavior, indicating better management of cuttings and entrapped air in drilling operations. Among the tested rheological models, the Herschel-Bulkley models offered the closest fluid behavior to the experimental data without and with the perlite addition. These findings highlight perlite’s potential as a valuable additive for enhancing water-based drilling fluids, enhancing operational efficiency and minimizing drilling complications. This study demonstrates perlite effectiveness as a primary additive in Micromax-weighted fluids, extending the operational density range and offering a practical, low-cost solution for HPHT conditions.

ZHANG JIN, ZHANG FENGQI, ZOU YANRONG et al.

Helium is a rare inert gas with indispensable applications in defense, aerospace, and medicine. However, helium resources available for use in China are extremely limited. To date, no independently accumulated helium resources have been found. Helium is primarily found in two forms: natural gas-associated and geothermal water-dissolved. This study focused on two typical basins—the Hangjinqi area in the northern Ordos Basin and the Weihe Basin—to investigate the genesis of helium. Helium isotope mass spectrometry analysis, rock radioactive element analysis and other methods were conducted to test the assgciated gas, core samples and potential helium source rock in the surrounding areas of the study area. The results show that helium in the Hangjinqi area in the northern Ordos Basin is typically crust-derived. While in Weihe Basin, high volume fractions of mantle-derived helium (up to 6.959%) were detected near deep-seated faults penetrating the basement, such as the Baoji-Xianyang fault and the Piedmont Fault of the Qinling Mountains. Both basins are located on the southwestern margin of the North China Plate and share a basement composed mainly of Archean-Proterozoic metamorphic-granite complexes, which serve as the main source rocks of helium formation. In addition, the main source rocks for helium gas in the Weihe Basin also include the uranium-rich granites of the Yanshanian period around the periphery and the concealed granitic bodies of the same period in the deep part of the basin. Due to the low mass fractions of U and Th elements or the low helium gas content of the desorbed gas in the basement sedimentary rock complexes, they cannot be regarded as the main source rocks for helium gas. The formation, migration and accumulation of helium gas in both areas are controlled by the source rocks and faults, and are closely related to the distribution of deep-seated fault zones. These findings provide a scientific basis for the further exploration and development of helium resources in the Weihe and northern Ordos basins.

Jiaxuan HU, Changqing GUO, Zhida WANG et al.

During the process of water electrolysis,the "bubble effect" will significantly reduce the overall performance of the system.The classical nucleation theory (CNT model） fails to reveal the regulatory mechanism of the electrical double layer (EDL),surface microstructure,and mass transfer synergy on nucleation kinetics in actual electrochemical systems.This study develops an electrode interface bubble nucleation model with the synergistic effect of electrical double layer⁃mass transfer⁃surface microstructure,considering the synergistic regulation mechanism of ion migration diffusion behavior,electrode surface nano microstructure,and concentration boundary layer on the nucleation process.The research results show that the synergistic effect of EDL and microporous structurel generates significant potential gradients at the surface micropores,leading to an increase in local supersaturation and prioritizing bubble nucleation.At high overpotentials,the effect of the concentration boundary layer on nucleation energy barrier exhibits a nonlinear relationship.The thinner the concentration boundary layer is,the more significant the decreasing trend of the nucleation rate at high potential will be.The growth of bubbles is dominated by the net concentration flux near the three⁃phase contact line (TPCL),exhibiting a two⁃stage growth characteristic.The study provides a theoretical basis for optimizing the surface design of gas evolution electrodes.

Zou Yiqi Chen Xuyue, Gao Deli

The new hybrid bit with central-grooved single-cone & multi-blade combines the advantages of PDC bit and single-cone bit， and effectively improves the rock-breaking efficiency in deep and ultra-deep well drilling. However， the concave-type structure and multi-nozzle layout make the downhole flow field more complex and the erosion wear phenomenon more remarkable. To reveal the erosion wear mechanism， the coupled computational fluid dynamics （CFD） and discrete phase model （DPM） method was employed to establish a numerical model of downhole flow field of the concave-type single-cone-PDC hybrid bit. The effects of drilling fluid discharge rate， cuttings diameter and cuttings mass flow rate on erosion wear were investigated， and fitting formulas for different factors on erosion rate were proposed. The study results show that increasing the drilling fluid discharge rate significantly enhances the jet flow kinetic energy， causing the erosion rate to rise from 5.0×10^-7 kg/（m²·s） to 19.5×10^-7 kg/（m²·s）， with erosion hotspots concentrated at the nozzle outlet and concave center region. Larger cuttings diameter increases particle kinetic energy and inertia， leading to high-frequency cyclic impacts at the nozzle outlet， blade front and concave center region. Higher cuttings mass flow rate notably elevates particle impact frequency， resulting in an upward trend in erosion rate and intensified erosion wear at the concave center region and nozzle outlet. The study results provide a theoretical basis for the hydraulic structure optimization and anti-erosion design of the hybrid bit with central-grooved single-cone & multi-blade.

LIU Xingbin, LI Shanwen, DENG Yuheng et al.

The correct selection of water nozzle diameter plays a crucial role in the accuracy of layered flow regulation during water injection construction operations. However, the selection of the diameter of the fishing type water injection nozzle mainly relies on manual experience, which leads to low construction efficiency and a long deployment cycles. Set up an experimental device to collect flow rate and pressure difference data at both ends of the water nozzle, and obtain a pressure difference and flow rate relationship chart from small diameter to large diameter water nozzle diameter range through linear fitting of experimental data. Based on the energy equation of fluid mechanics water nozzle throttling, establish a formula for the relationship between local resistance coefficient and water nozzle area for different diameter water nozzle throttling. According to this formula, the water nozzle diameter under given flow rate and pressure difference conditions can be accurately calculated. Compare the on-site experimental data with the formula validation data and the literature experimental data with the formula validation data, and the error is no more than 5%. Establish three-dimensional models of water nozzles with different diameters, use FLUENT software for numerical simulation to calculate the pressure difference data of water nozzles with different diameters at diferent flow rates. Compare the flow pressure difference data obtained from numerical simulation with experimental data, the simulation results is no more than 12% between the experimental and simulation results. The numerical simulation error is within the range of the engineering accuracy requirements, and further verifing the reliability of the formula for the relationship between the local resistance coefficient of the water nozzle and the water nozzle area. The use of nozzle diagrams and relationship formulas is beneficial for accurate selection of nozzles in practical field applications, and has important value in improving the efficiency and accuracy of layered water injection flow regulation in oil fields.

Tuyen Tran, Thao Minh Le, Quang-Hung Le et al.

Vision-language alignment in video must address the complexity of language, evolving interacting entities, their action chains, and semantic gaps between language and vision. This work introduces Planner-Refiner, a framework to overcome these challenges. Planner-Refiner bridges the semantic gap by iteratively refining visual elements' space-time representation, guided by language until semantic gaps are minimal. A Planner module schedules language guidance by decomposing complex linguistic prompts into short sentence chains. The Refiner processes each short sentence, a noun-phrase and verb-phrase pair, to direct visual tokens' self-attention across space then time, achieving efficient single-step refinement. A recurrent system chains these steps, maintaining refined visual token representations. The final representation feeds into task-specific heads for alignment generation. We demonstrate Planner-Refiner's effectiveness on two video-language alignment tasks: Referring Video Object Segmentation and Temporal Grounding with varying language complexity. We further introduce a new MeViS-X benchmark to assess models' capability with long queries. Superior performance versus state-of-the-art methods on these benchmarks shows the approach's potential, especially for complex prompts.

Huaixia Dou, Xinyu Tian, Xinglin Lyu et al.

Recent advancements in large language models (LLMs) have demonstrated their remarkable capabilities across various language tasks. Inspired by the success of text-to-text translation refinement, this paper investigates how LLMs can improve the performance of speech translation by introducing a joint refinement process. Through the joint refinement of speech translation (ST) and automatic speech recognition (ASR) transcription via LLMs, the performance of the ST model is significantly improved in both training-free in-context learning and parameter-efficient fine-tuning scenarios. Additionally, we explore the effect of document-level context on refinement under the context-aware fine-tuning scenario. Experimental results on the MuST-C and CoVoST 2 datasets, which include seven translation tasks, demonstrate the effectiveness of the proposed approach using several popular LLMs including GPT-3.5-turbo, LLaMA3-8B, and Mistral-12B. Further analysis further suggests that jointly refining both transcription and translation yields better performance compared to refining translation alone. Meanwhile, incorporating document-level context significantly enhances refinement performance. We release our code and datasets on GitHub.

Qibin Wang, Pu Zhao, Shaohan Huang et al.

To further enhance the ability of Large Language Models (LLMs) to solve complex, multi-step reasoning problems, test-time scaling (TTS) methods have gained widespread attention. Existing approaches such as Best-of-N and majority voting are limited as their performance depends on the quality of candidate responses, making them unable to produce a correct solution when all candidates are incorrect. Introducing an additional model to select the best response also incurs significant deployment costs. To this end, we introduce Generative Self-Refinement (GSR), a novel parallel test-time scaling framework where a unified model first generates a set of candidate responses in parallel and then performs self-refinement to synthesize a new superior solution based on a prompt consisting of the problem and these candidates. However, LLMs struggle to perform refinement effectively when prompted directly. Therefore, we design a hybrid training pipeline by jointly optimizing for two complementary objectives, solving problems directly and refining candidate responses. Experimental results demonstrate that our method achieves state-of-the-art performance across five mathematical benchmarks. We further show that this learned self-refinement skill is a model-agnostic enhancement, robust across different model scales and generalizing to out-of-distribution reasoning tasks.

WANG Jie, WANG Qianjun, ZHENG Sheng et al.

Due to large burial depths， ultra-deep clastic rock reservoirs have poor physical properties and low oil and gas productivity. However， the newly drilled Well Zheng10 in Shawan Sag， Sinopec exploration area， has encountered high-quality thick reservoirs in the Triassic Karamay Formation at a buried depth of 6 700 m， and conventional tests have obtained high industrial oil and gas flow. To reveal the characteristics of this set of reservoirs and favorable main controlling factors， clarify the direction of ultra-deep clastic rock exploration， and reduce the risk of further oil and gas exploration in this area， this paper comprehensively analyzed the data of core， well logging， cast thin sections， physical properties， scanning electron microscopy， and diagenetic evolution of the reservoirs， and discussed the petrological， physical， and pore characteristics of reservoirs in the Triassic Karamay Formation in this area. The results show that the ultra-deep clastic rocks of the Triassic Karamay Formation in Well Zheng10 area are deposited in the front of the braided river delta. The lithology is dominated by glutenites， gravel-bearing fine sandstones， and fine sandstones， and the rock types are mainly feldspar lithic sandstones with low composition maturity. The remaining primary pores and solution pores dominate the reservoir space. The average porosity of the reservoirs is 9.1%， and the average permeability is 2.85 mD. The reservoir belongs to the ultra-low porosity and ultra-low permeability type and low porosity and low permeability type as a whole. Favorable sedimentary facies zones， constructive diagenesis， and abnormal high-pressure control the development and distribution of favorable reservoirs in this area. Among them， sedimentary microfacies are the basis of favorable reservoir development. The fine sandstones deposited in the underwater distributary channel at the front of the braided river delta are better sorted and rounded. The contents of the muddy matrix are low， and the physical properties of the reservoir are better， which are the dominant phase zones for favorable reservoir development. The sodium feldspar cements generated in the early diagenetic process are dissolved by acids in the later period， which increases the dissolution pores and is an essential constructive diagenesis. The abnormal high-pressures formed by continuous oil and gas charging form early and develop continuously for a long time， playing an important construction role. The research results have a guiding role for the subsequent ultra-deep oil and gas exploration in this area.

Lora R. Du, Kathy Q. Ji

The note is dedicated to refining a theorem by Diaconis, Evans, and Graham concerning successions and fixed points of permutations. This refinement specifically addresses non-adjacent successions, predecessors, excedances, and drops of permutations.

Silvia Frassu, Giuseppe Viglialoro

These notes aim to provide a deeper insight on the specifics of the paper "Refined criteria toward boundedness in an attraction-repulsion chemotaxis system with nonlinear productions" by A. Columbu, S. Frassu and G. Viglialoro [\textit{Appl. Anal.} 2024, 103:2, 415--431].

CHEN Minfeng,QIN Lifeng,ZHAO Kang,WANG Yiwen

For designing effective injection and production well spacing in low permeability reservoirs, it's essential to consider the pressure-sensitive effects arising from various pressure changes during the operation of injection and production wells, along with the impact of changes in the start-up pressure gradient. This study builds upon the basic seepage laws of low-permeability reservoirs to establish a seepage equation that incorporates the effects of the starting pressure gradient and pressure sensitivity. Utilizing the stable successive substitution method, the study examines the pressure distribution and the mechanism of reserve production under typical injection-production patterns. Based on the different requirements of the daily output of oil wells in an actual oil fields, a method for solving the effective injection-production well spacing of low-permeability pressure sensitive reservoir is determined. The study shows that, using the pressure sensitivity of the injection and production well area as a benchmark, there is a notable difference in the calculated results for well spacing when compared with scenarios where pressure sensitivity is either not considered or only considered for the production well area. Specifically, these differences are +9.8 % and -20.6 % under the same conditions. Considering the production limit requirements, the effective injection-production well spacing is about 0.7 ~ 0.9 times of the limit injection-production well spacing under normal conditions, which can better guide the reasonable deployment of the development well pattern of low permeability pressure sensitive reservoirs.

David A. Wood

Derivative/volatility well-log attributes from very few commonly recorded well logs can assist in the prediction of total organic carbon (TOC) in shales and tight formations. This is of value where only limited suites of well logs are recorded, and few laboratory measurements of TOC are conducted on rock samples. Data from two Lower-Barnett-Shale (LBS) wells (USA), including well logs and core analysis is considered. It demonstrates how well-log attributes can be exploited with machine learning (ML) to generate accurate TOC predictions. Six attributes are calculated for gamma-ray (GR), bulk-density (PB) and compressional-sonic (DT) logs. Used in combination with just one of those recorded logs, those attributes deliver more accurate TOC predictions with ML models than using all three recorded logs. When used in combination with two or three of the recorded logs, the attributes generate TOC prediction accuracy comparable with ML models using five recorded well logs. Multi-K-fold-cross-validation analysis reveals that the K-nearest-neighbour algorithm yields the most accurate TOC predictions for the LBS dataset. The extreme-gradient-boosting (XGB) algorithm also performs well. XGB is able to provide information about the relative importance of each well-log attribute used as an input variable. This facilitates feature selection making it possible to reduce the number of attributes required to generate accurate TOC predictions from just two or three recorded well logs.

Khalismatov I.Kh., Makhmudov N.N., Zakirov R.T. et al.

Depending on the peculiarities of the geological, hydrogeological, technical and economic conditions for the development of gas and gas condensate fields in the Ustyurt petroleum region, the final gas recovery from terrigenous Jurassic reservoirs can vary widely. The increase in the final gas recovery is one of the main issues of the rational development of gas and gas condensate fields and depends on geological and technological factors.

Li Teng, Sun Zhiqian, Wang Chaolei et al.

In order to study the internal flow field and separation characteristics of micro double volute gas-liquid cyclone separator,a physical model of the separator was established by Fluent software.Based on numerical simulations for the distribution of gas flow field and the trajectory of droplet particles,the influences of droplet size and inlet velocity on the separation efficiency were investigated.It is found that short circuit phenomenon is obvious near the inlet of exhaust pipe,the tangential velocity and axial velocity are distributed uniformly along the radial direction,radial velocity is of small order of magnitude with complex distribution,and the static pressure and dynamic pressure are distributed symmetrically.In the process of gas-liquid two-phase flow,most droplets move to the side wall of the separator under the action of centrifugal force to be separated,and some droplet particles escape upwards via the exhaust pipe due to short circuit and back mixing.Within the scope of the study,larger droplet size contributes to higher separation efficiency.The critical droplet size of the separator is 1.0 μm.For droplets equal to or larger than 1.0 μm,the separation efficiency increases with the increase of inlet velocity; for droplets smaller than 1.0 μm,the separation efficiency increases first and then decreases with the increase of inlet velocity and the highest separation efficiency could be reached if proper inlet velocity is selected.The study results provide reference for the development of high-efficiency micro gas-liquid cyclone separators.

Huanchen Bao, Xuhua He

We show that the totally nonnegative part of the twisted product of flag varieties of a Kac-Moody group admits a cellular decomposition, and the closure of each cell is a topological manifold with boundary. We also establish explicit parameterizations of each totally positive cell. In the special cases of double flag varieties and braid varieties, we show that the totally nonnegative parts are regular CW complexes homeomorphic to closed balls. Moreover, we prove that the link of any totally nonnegative double Bruhat cell in a reductive group is a regular CW complex homeomorphic to a closed ball, solving an open problem of Fomin and Zelevinsky.

Li Zhao, Chunsheng Feng, Chensong Zhang et al.

The black oil model is widely used to describe multiphase porous media flow in the petroleum industry. The fully implicit method features strong stability and weak constraints on time step-sizes; hence, commonly used in the current mainstream commercial reservoir simulators. In this paper, a CPR-type preconditioner with an adaptive "setup phase" is developed to improve parallel efficiency of petroleum reservoir simulation. Furthermore, we propose a multi-color Gauss-Seidel (GS) algorithm for algebraic multigrid method based on the coefficient matrix of strong connections. Numerical experiments show that the proposed preconditioner can improve the parallel performance for both OpenMP and CUDA implements. Moreover, the proposed algorithm yields good parallel speedup as well as same convergence behavior as the corresponding single-threaded algorithm. In particular, for a three-phase benchmark problem, the parallel speedup of the OpenMP version is over 6.5 with 16 threads and the CUDA version reaches more than 9.5.

Jureeporn Yuennan, Phongpichit Channuie

A natural combination of the first and second derivatives of the scalar potential was achieved in a framework of an alternative refined de Sitter conjecture recently proposed in the literature. In this work, we study various inflation models in which the inflaton is a composite field emerging from various strongly interacting field theories. We then examine if these three models of inflation can satisfy this further refining de Sitter swampland conjecture or not. Regarding our analysis with proper choices of parameters $a,\,b=1- a$ and $q$, we find that some inflationary models are in strong tension with the refined Swampland conjecture. However, all of them can always satisfy the alternative refined de Sitter conjecture. Therefore, one may expect that all inflationary models might all be in landscape since the further refining de Sitter swampland conjecture is satisfied.

Bryan Zhu, Nicholas Lui, Jeremy Irvin et al.

Reducing methane emissions is essential for mitigating global warming. To attribute methane emissions to their sources, a comprehensive dataset of methane source infrastructure is necessary. Recent advancements with deep learning on remotely sensed imagery have the potential to identify the locations and characteristics of methane sources, but there is a substantial lack of publicly available data to enable machine learning researchers and practitioners to build automated mapping approaches. To help fill this gap, we construct a multi-sensor dataset called METER-ML containing 86,599 georeferenced NAIP, Sentinel-1, and Sentinel-2 images in the U.S. labeled for the presence or absence of methane source facilities including concentrated animal feeding operations, coal mines, landfills, natural gas processing plants, oil refineries and petroleum terminals, and wastewater treatment plants. We experiment with a variety of models that leverage different spatial resolutions, spatial footprints, image products, and spectral bands. We find that our best model achieves an area under the precision recall curve of 0.915 for identifying concentrated animal feeding operations and 0.821 for oil refineries and petroleum terminals on an expert-labeled test set, suggesting the potential for large-scale mapping. We make METER-ML freely available at https://stanfordmlgroup.github.io/projects/meter-ml/ to support future work on automated methane source mapping.

Alessandro Daniele, Emile van Krieken, Luciano Serafini et al.

Recent work has shown logical background knowledge can be used in learning systems to compensate for a lack of labeled training data. Many methods work by creating a loss function that encodes this knowledge. However, often the logic is discarded after training, even if it is still useful at test time. Instead, we ensure neural network predictions satisfy the knowledge by refining the predictions with an extra computation step. We introduce differentiable refinement functions that find a corrected prediction close to the original prediction. We study how to effectively and efficiently compute these refinement functions. Using a new algorithm called Iterative Local Refinement (ILR), we combine refinement functions to find refined predictions for logical formulas of any complexity. ILR finds refinements on complex SAT formulas in significantly fewer iterations and frequently finds solutions where gradient descent can not. Finally, ILR produces competitive results in the MNIST addition task.