检索结果-内蒙古大学图书馆

Interpretable Metric Learning: A Weight distribution Optimization Perspective

Procedia Computer Science 2024年 247卷 93-102页

作者： Xinyue Miao Le Xu Zhang Wen Lei Zhang Zhiqiang Wei Changtao Wu Zhengkai Shang Kelin Wang Qianxing deng Fuyuan Wei Liyan Liu R&D Center for Ultra Deep Complex Reservior Exploration and Development CNPC 841000 Xinjiang China Network and Information Center Southwest Petroleum University Chengdu 610500 Sichuan Province China Tarim Oilfield Branch Company Petrochina 841000 Xinjiang China School of Computer Science and Software Engineering Southwest Petroleum University Chengdu China

Metric learning aims to reduce intra-class distance and expand inter-class distance. This paper proposes a distribution optimization algorithm that incorporates interpretable feature engineering (dOIFE). dOIFE explores the local interpretability of each instance, reconstructs the feature space of the training set with reverse weighting, and optimizes the feature space of the target domain to address metric issues. Our approach innovatively derives interpretability weights from the classifier's perspective and applies fine-grained weighting to the instances to optimize the spatial metric. For any classifier, we conduct local interpretability analysis on its results to obtain fine-grained sample weights. In the training space, we designed an interpretable feature engineering method to expand inter-class distance and reduce intra-class distance. For the test space, we developed a unified method that combines ensemble inference and distance metrics to optimize feature distribution. We conducted experiments on classical balanced datasets, long-tail datasets, and image datasets. Comparative experiments with nine distance metric learning algorithms demonstrated the superiority of our algorithm. Statistical tests show that our algorithm can be applied to various classification models, improve classification performance, and offer a certain level of interpretability.

关键词： Metrics Learning Classification Algorithm

来源：评论

学校读者我要写书评

暂无评论

Construction and Performance Study of Quaternary Ammonium Salt Inhibitor for Carbon Steel Corrosion in High Temperature (180°C) during Oilfield Acidification

SSRN

引用

SSrN 2024年

作者： Li, Xianghong Zhang, Fengrunze Liu, Peng Li, Hua Yu, Weiming Zhong, Hailian Liu, Shanghao Wang, Jiaqin Bao, dan Tan, Bochuan Zhang, Peng Chongqing University of Science and Technology School of Chemistry and Chemical Engineering Chongqing401331 China R&D Center China Petroleum Engineering & Construction Corp. North China Company Hebei Renqiu062550 China

AbstractsSynthesis of quaternary ammonium salt inhibitor (1,4-phenylethyl (methylene)) bis (1,10-phenanthroline-1-xuan-1-yl) chloride (I)) (PdQ) from 1,10-phenanthroline and 1,4-bis (chloromethyl) benzene. Corrosion inhibition performance and corrosion inhibition mechanism of PdQ in 20% HCl for N80 steel sheet by weightlessness method, electrochemistry, scanning electron microscopy and theoretical calculations. The results showed that the corrosion rate was 69.52 g/(m2·h) at 180℃, 3 MPa and 20% HCl for 4 h, which met the industry standard. Electrochemical tests indicate that PdQ is an anodic corrosion inhibitor and that adsorption on the N80 surface is a geometric coverage effect. SEM allows observation of the surface condition of metals under different conditions. From quantum chemical calculations, it can be seen that PdQ mainly acts as an electron acceptor and forms adsorption by accepting d-orbital electrons from Fe atoms on the metal surface. © 2024, The Authors. All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

The Multiphase Flow rate Measurement Model of Non-Full-Pipe Based on Fluid Mechanics Coupled with Iscn-Lstm Network

SSRN

引用

SSrN 2022年

作者： Liang, Haibo Song, Chuanhao Wang, ren School of Mechanical Engineering Southwest Petroleum University Chengdu610500 China CNPC Engineering Technology R & D Company Limited Beijing102206 China

Based on the theory of ultrasonic doppler frequency shift extraction and combined with pipeline fluid mechanics, the superficial velocity model of Non-Full-Pipe is established by using four-channel ultrasonic array. Subsequently, ISCN-LSTM network training was carried out after ultrasonic signal data fusion to obtain the liquid level identification model. By introducing the theory of Velocity-Liquid-Level, a new multiphase flow calculation model for Non-Full-Pipe is established to calculate the overall superficial flow rate of it. The measurement experiment of Non-Full-Pipe was carried out to verify the performance of the new model. The experimental results show that compared with the CFMS composed of multiphase target flowmeters with ± 0.5 % FS, the accuracy of the UFMS using the new model is reduced by 0.9645 %, the nonlinear error is reduced by 2.293 %, the ArE is reduced by 2.57 %, the Sd is reduced by 1.39578, and the rMSE is reduced by 14.39326. © 2022, The Authors. All rights reserved.

关键词： data fusion

来源：评论

学校读者我要写书评

暂无评论

Mechanistic Insights into Janus Silica Nanoparticles as Shale Inhibitors in Water-Based drilling Fluids

SSRN

引用

SSrN 2024年

作者： Lv, Kaihe dai, Jiajun Sun, Jinsheng Jia, Han Zhang, Chao Huang, Xianbin Shang, Jundi Qingdao266580 China Ministry of Education Qingdao266580 China CNPC Engineering Technology R & D Company Limited Beijing102206 China Key Laboratory of Colloid and Interface Chemistry Shandong University Ministry of Education Jinan250100 China

Shale hydration expansion is a serious challenge in current drilling fluid technology, which poses significant hindrances to efficient oil and gas extraction. This study investigated the potential of Janus silica nanoparticles (JNP) as novel shale inhibitors. JNP with both hydrophilic and hydrophobic characteristics was synthesized via the Pickering emulsion method. Its inhibitive properties were evaluated and compared with traditional inhibitors, namely KCl, PA, and SiO2, by the bentonite inhibition test, linear swelling test, and shale cuttings hot-rolling recovery test. Notably, at just a much low concentration of 0.5 wt%, JNP reduced the linear swelling rate from 103.1% to 40.2% and achieved a shale cutting recovery rate of 60.7% at 160℃. Based on contact angle measurements, scanning electron microscopy (SEM) observations, molecular dynamics simulation, ζ-potential measurements, X-ray diffraction (Xrd) analysis, and liquid level and capillary pressure measurements, it was proposed that the superior inhibitive performance of JNP arose from its ability to compress the electrical layer, block nanopores, and alter wettability. The novel JNP exhibited considerable potential as an effective inhibitor in water-based drilling fluids for shale oil and gas drilling. © 2024, The Authors. All rights reserved.

关键词： Emulsification

来源：评论

学校读者我要写书评

暂无评论

IoT-driven defect detection System for Plastic recycling Plants Based on Convolutional Neural Networks 2

IoT-Driven Defect Detection System for Plastic Recycling Pla...

引用

2nd International Conference on Self Sustainable Artificial Intelligence Systems, ICSSAS 2024

作者： Madhiraju, Priyanka Sampathrajan, rajeshkumar Pattnaik, Manjula rajalakshmi, P. dhivya, K. Murugan, S. Matrusri Engineering College Department of Computer Science and Engineering Telangana Hyderabad India McKinsey & Company Fort WorthTX United States Princess Nourah Bint Abdulrahman University College of Business and administration Accounting Department Saudi Arabia Vel Tech Rangarajan Dr.Sagunthala R&d Institute of Science and Technology Department of Computer Science and Engineering Tami Nadu Chennai India Srm Institute of Science and Technology Department of Electronics and Communication Engineering Kattankulathur Tamil Nadu Chennai India Saveetha Institute of Medical and Technical Sciences Saveetha University Saveetha School of Engineering Department of Biomedical Engineering Tamil Nadu Chennai India

ISBN: (纸本)9798350368413

To improve the accuracy of plastic recycling facilities this paper introduces an Internet of Things (IoT)driven defect detection system that uses Convolutional Neural Networks (CNNs). Traditional techniques of fault identification are time-consuming and prone to human error, which makes plastic recycling less than ideal from an environmental sustainability perspective. To track the movement of plastics in real time, the system makes use of IoT sensors placed all throughout the recycling facility. To correctly detect plastic material flaws like contamination, discoloration, or physical damage, this data is then input into a CNNs model that has been tuned for image analysis. The proposed system outperforms conventional approaches in terms of efficiency and accuracy of flaw identification, as shown by rigorous testing. This method improves the quality and longevity of plastic recycling operations while simultaneously streamlining defect identification procedures via the use of IoT technology and CNNs. The IoT presents exciting new methods to lessen the negative effects of plastic recycling facilities on the environment while simultaneously increasing their efficiency and effectiveness, meeting the growing demand for environmentally friendly waste management solutions throughout the world. © 2024 IEEE.

关键词： Plastics plants

来源：评论

学校读者我要写书评

暂无评论

Assessing Two-Well-Horizontal Caverns for Compressed Air Energy Storage in Low-Grade Salt rocks

SSRN

引用

SSrN 2024年

作者： Li, Qihang Liu, Wei Jiang, Liangliang Qin, Jianxu Wan, Jifang Zhu, Xuanshi State Key Laboratory of Coal Mine Disasters Dynamics and Control Chongqing University Chongqing400044 China State Key Laboratory of Geomechanics and Geotechnical Engineering Institute of Rock and Soil Mechanics Chinese Academy of Sciences Wuhan430071 China Department of Chemical and Petroleum Engineering the University of Calgary CalgaryAB Canada CNPC Engineering Technology R&D Company Limited Beijing102206 China

To advance the development and deployment of compressed air energy storage (CAES) in underground salt caverns (USCs) for a carbon-resilient society, a comprehensive stability assessment is essential to ensure the safe operation of underground salt cavern gas storage (SCGS). In China, around 85% of low-grade salt rock features two-well-horizontal (TWH) caverns, which are notable for their high sediment content (SC) and restricted usable space. This study addresses the challenges associated with sedimentary voids energy storage and the stability of TWH-caverns for large-scale energy storage (LSES). We present a theoretical model for the volumetric inversion of TWH-caverns, based on salt extraction, salt layer characteristics, and interlayer ratio, using data from a low-grade salt mine in Feicheng City, China. We simulated the volume loss rate (VLr), displacement, and plastic zone volume (PZV) for various levels of high SC (50%, 60%, 70%, 80%, 90%) and internal air pressure (IAP) (8.1 MPa, 9.45 MPa, 21.6 MPa) in the roof, waist, and connected areas of both high and low caverns within the TWH-cavern. The results reveal that the creep characteristics of the high cavern are more pronounced, and the combined effects of SC and IAP effectively limit roof deformation. However, the displacement of the high cavern waist increases by 0.028 m (IAP = 8.1 MPa) and 0.006 m (IAP = 21.6 MPa) as SC rises from 80% to 90%. In the connected area, displacement values increase from 0.634 m to 0.852 m (IAP = 8.1 MPa) as SC increases from 50% to 90%. These findings highlight that the sediment surface is a critical factor in assessing the impact of SC on cavern deformation. The cavern roof remains the most critical component for ensuring the safe operation of SC- CAES. We propose effective measures to enhance roof stability and suggest establishing an intelligent big data platform to ensure the secure operation of low-grade SC-CAES. Accordingly, this study provides a theoretical basis for eval

关键词： digital storage

来源：评论

学校读者我要写书评

暂无评论

Origin of gas condensate reservoir in Fuman Oilfield, Tarim Basin, NW China

引用

Petroleum Exploration and development 2023年第6期50卷 1295-1307页

作者： WANG Qinghua Tarim Oilfield Company PetroChinaKorlaXinjiang 841000China R&D Center for Ultra-Deep Complex Reservoir Exploration and Development CNPCKorlaXinjiang 841000China Engineering Research Center of Ultra-deep Complex Oil and Gas Reservoir Exploration and Development Xinjiang Uygur Autonomous RegionKorlaXinjiang 841000China

To understand the reservoir property and hydrocarbon accumulation conditions of the Middle and Upper Ordovician intraplatform shoal between ultra-deep main strike-slip faults in Fuman Oilfield of the Tarim Basin, China, the main strike-slip faults in and around well Fd1 in the basin were analyzed in terms of sedimentary facies, sequence stratigraphy, intraplatform shoal reservoir property, and oil and gas origins, based on drilling data. The Yingshan Formation intraplatform shoal between the main strike-slip faults is superimposed with low-order faults to form a new type of hydrocarbon play. Firstly, hydrocarbons generated from the Lower Cambrian Yuertusi Formation source rocks vertically migrated into the second member of Yingshan Formation through the main strike-slip faults, and then migrated laterally until they were accumulated. A small amount of oil from Well Fd1 came from the Yuertusi Formation source rocks in the mature stage, and a large amount of gas originated from oil cracking in the ultra-deep reservoirs. Therefore, the secondary gas condensate reservoir in Well Fd1 is characterized by high gas to oil ratio, dry gas (dryness coefficient being 0.970) and hybrid origin. This new type of hydrocarbon play characterized by intraplatform shoal and low-order fault suggests a prospect of continuous hydrocarbon-bearing area in Fuman Oilfield, which will expand the ultrap-deep oil and gas exploration in the oilfield.

关键词： low-order fault intraplatform shoal ultra-deep Yingshan Formation oil cracked gas condensate gas Fuman Oilfield Tarim Basin

来源：评论

学校读者我要写书评

暂无评论

Comprehensive Safety Assessment Of Two-Well-Horizontal Caverns with Sediment Space For Compressed Air Energy Storage In Low-Grade Salt rocks

SSRN

引用

SSrN 2024年

作者： Li, Qihang Liu, Wei Jiang, Liangliang Qin, Jianxu Wan, Jifang Zhu, Xuanshi State Key Laboratory of Coal Mine Disasters Dynamics and Control Chongqing University Chongqing400044 China State Key Laboratory of Geomechanics and Geotechnical Engineering Institute of Rock and Soil Mechanics Chinese Academy of Sciences Wuhan430071 China Department of Chemical and Petroleum Engineering The University of Calgary CalgaryAB Canada CNPC Engineering Technology R&D Company Limited Beijing102206 China

To expedite the construction and implementation of compressed air energy storage (CAES) in under- ground salt caverns (USCs), conducting a thorough stability assessment is crucial to ensure the safe operation of underground salt cavern gas storage (SCGS). In China, around 85% of low-grade salt rock features two-well-horizontal (TWH) caverns, which are notable for their high sediment content (SC) and restricted usable space. Herein this study addresses the challenges associated with sediment voids energy storage and the stability of TWH-caverns for large-scale energy storage (LSES). We present a theoretical model for the volumetric inversion of TWH-caverns, based on salt extraction, salt layer characteristics, and interlayer ratio, using data from a low-grade salt mine in Feicheng City, China. We simulated the volume loss rate (VLr), displacement, and plastic zone volume (PZV) for various levels of high SC (50%, 60%, 70%, 80%, 90%) and internal air pressure (IAP) (8.1 MPa, 9.45 MPa, 21.6 MPa) in the roof, waist, and connected areas of both high and low caverns within the TWH-cavern. The results reveal that the creep characteristics of the high cavern are more pronounced, and the combined effects of SC and IAP effectively limit roof deformation. However, the displacement of the high cavern waist increases by 0.028 m (IAP = 8.1 MPa) and 0.006 m (IAP = 21.6 MPa) as SC rises from 80% to 90%. In the connected area, displacement values increase from 0.634 m to 0.852 m (IAP = 8.1 MPa) as SC increases from 50% to 90%. These findings highlight that the sediment surface is a critical factor in assessing the impact of SC on cavern deformation. The cavern roof remains the most critical component for ensuring the safe operation of SC-CAES. We propose effective measures to enhance roof stability and suggest establishing an intelligent big data platform to ensure the secure operation of low-grade SC-CAES. Accordingly, this study provides a theoretical basis for evaluating the safet

关键词： digital storage

来源：评论

学校读者我要写书评

暂无评论

Field stabilization of pulse duration in a hundredfemtosecond level

引用

Chinese Optics Letters 2024年第8期22卷 93-99页

作者： runmin Liu Yong Wu Guoqing Pu Jiayang Cheng Huan Mu Bowen Liu Lilin Yi Minglie Hu Ultrafast Laser Laboratory Key Laboratory of Opto-electronic Information Science and Technology of Ministry of EducationSchool of Precision Instruments and Opto-electronics EngineeringTianjin UniversityTianjin 300072China State Key Laboratory of Advanced Communication Systems and Networks School of Electronic Information and Electrical EngineeringShanghai Jiao Tong UniversityShanghai 200240China R&D Center AIOPTICS Company LimitedShanghai 200240China

Pulse duration is considered as one of the most important characteristics of high-power femtosecond ***,pulses output from the laser system are susceptible to ambient changes and manifest the instability of pulse durations in an open *** this paper,incorporating the algorithmic framework of the improved stochastic hill-climbing search and incremental proportional-integral controller,temperature-induced fluctuations of pulse duration can be effectively compensated by an automatic feedback control in an all-fiber chirped-pulse amplification *** the experiment,sub-hundred femtosecond fluctuation of pulse duration is introduced to verify the performance robustness of the proposed pulse-duration feedback control(PdFC).The stability of pulse duration is obviously higher than the case without the feedback control,and the peak-to-peak fluctuation of pulse duration is reduced to 6.5%.Furthermore,the robust switching between different pulse durations proves the versatility of the *** expect that the proposed feedback control method could provide a novel insight into high-power femtosecond lasers widely applied in fundamental researches and industrial fields.

关键词： fiber amplifier femtosecond laser pulse-duration feedback control

来源：评论

学校读者我要写书评

暂无评论

In-Situ Catalytic Pyrolysis of Oil Shale Using Cu@Zeolite Catalyst and Kinetics Analysis

SSRN

引用

SSrN 2022年

作者： Jin, Jiafeng Sun, Jinsheng Lv, Kaihe Hou, Qilin Guo, Xuan Liu, Kesong Ministry of Education Shandong Qingdao266580 China Qingdao China CNPC Engineering Technology R & D Company Limited Beijing102206 China

The in-situ catalytic conversion technology has been regarded as a versatile approach to developing oil shale due to the advantages of environmental protection and being able to exploit the deep formations, while the high conversion temperature and low oil recovery are the key deficiencies that hinder the progress of the commercialized operation. In this study, a novel Cu@4A zeolite nano-catalyst for oil shale in-situ conversion was synthesized by the combination of the wet impregnation method and high-temperature calcination. The catalyst can enter the pore throat and pyrolyze the kerogen at a lower temperature, exhibiting favorable catalytic performance toward oil shale. The functional groups, micro-morphology, crystal structure information, elements valence state, and the thermostability of the prepared catalyst were characterized through FT-Ir, SEM, Xrd, XPS, and TGA techniques, results are consistent with the expected design. The effects of Cu@4A zeolite nano-catalyst on the hydrocarbon yields and kinetics were investigated, and results show that the oil and gas yields increased by 7.61% and 19.69% after adding 1% Cu@4A catalyst;the activation energy of the oil shale sample decreased from 190.73 to 147.9512 KJ/mol, and the hydrocarbon conversion is proportional to the mesh number of oil shale, the optimal heating rate is 10 °C/min. The possible explanation for its favorable catalytic performance can be that the kerogen and asphalt in oil shale would firstly decompose into macromolecule aromatic hydrocarbons via the β-cleavage of the carbenium ions over the Brønsted acidic sites of the prepared catalyst, then followed by the dehydrocyclization, oligomerization, and aromatization. The introduction of Cu can increase the quantity of Lewis acid sites on the framework of catalysts, leading to more hydrogen transfer reactions between catalyst and organics, achieving the high-efficiency catalysis of oil shale. © 2022, The Authors. All rights reserved.

关键词： Zeolites

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：