检索结果-内蒙古大学图书馆

A novel semi-analytical meshless method for Navier-Stokes equations arising in fractured rock masses

Chinese Journal of Physics 2025年 95卷 1069-1077页

作者： Jiang, Yingzi Wang, Fuzhang Institute of Data Science and Engineering Xuzhou University of Technology Xuzhou 221018 China

The seepage problem of oil-water two-phase flow in fractured rock masses has attracted great attention in engineering problems such as nuclear energy waste discharge. In this paper, we intend to propose a novel semi-analytical meshless method to simulate the Navier-Stokes equations arising in fractured rock masses. The semi-analytical meshless method is based on the newly-constructed Gaussian-type spatiotemporal radial basis functions. This method is a one-step solution method for time-dependent problems. Numerical examples show that the proposed semi-analytical meshless method can compete with the other methods in terms of solution accuracy as well as CPU time. © 2025 The Physical Society of the Republic of China (Taiwan)

关键词： Navier Stokes equation Radial basis functions Semi-analytical meshless method Time-dependent problems

来源：评论

学校读者我要写书评

暂无评论

Effect of Mn alloying on the hydrogen-assisted cracking behavior in multiphase/duplex stainless steel

引用

材料科学技术（英文版） 2025年第4期207卷 126-141页

作者： Menghao Liu Cuiwei Du Xiaogang Li Beijing Advanced Innovation Center for Materials Genome Engineering Institute for Advanced Materials and TechnologyUniversity of Science and Technology BeijingBeijing 100083China Key Laboratory for Corrosion and Protection of the Ministry of Education Institute of Advanced Materials & TechnologyUniversity of Science and Technology BeijingBeijing 100083China National Materials Corrosion and Protection Data Center University of Science and Technology BeijingBeijing 100083China

There remains debate on whether Mn is beneficial or detrimental to hydrogen embrittlement in stainless *** this work,a series of stainless steels were designed to study the change of hydrogen embrit-tlement sensitivity,crack propagation,and hydrogen trapping behaviors upon Mn *** results suggest that adding 4 wt.％Mn increased hydrogen embrittlement susceptibility,whereas adding 8 wt.％Mn decreased hydrogen embrittlement *** banded α'-martensite through austenitic grain is the main reason for the increased hydrogen embrittlement sensitivity when adding 4 wt.％Mn,by ad-sorbing hydrogen,promoting crack initiation,and accelerating crack propagation.

关键词： Hydrogen-assisted cracking Multiphase stainless steel Manganese

来源：评论

学校读者我要写书评

暂无评论

Integral experiment on slab^(nat)Pb using D-T and D-D neutron sources to validate evaluated nuclear data

引用

Nuclear science and Techniques 2025年第3期36卷 119-133页

作者： Kuo-Zhi Xu Yang-Bo Nie Chang-Lin Lan Yan-Yan Ding Shi-Yu Zhang Qi Zhao Xin-Yi Pan Jie Ren Xi-Chao Ruan Key Laboratory of Nuclear Data China Institute of Atomic EnergyBeijing 102413China School of Nuclear Science and Technology Lanzhou UniversityLanzhou 730000China

Lead(Pb)plays a significant role in the nuclear industry and is extensively used in radiation shielding,radiation protection,neutron moderation,radiation measurements,and various other critical ***,the measurement and evaluation of Pb nuclear data are highly regarded in nuclear scientific research,emphasizing its crucial role in the *** the time-of-flight(ToF)method,the neutron leakage spectra from three^(nat)Pb samples were measured at 60°and 120°based on the neutronics integral experimental facility at the China institute of Atomic Energy(CIAE).The^(nat)Pb sample sizes were30 cm×30 cm×5 cm,30 cm×30 cm×10 cm,and 30 cm×30 cm×15 *** sources were generated by the Cockcroft-Walton accelerator,producing approximately 14.5 MeV and 3.5 MeV neutrons through the T(d,n)^(4)He and D(d,n)^(3)He reactions,*** neutron spectra were also calculated by employing the Monte Carlo code of MCNP-4C,and the nuclear data of Pb isotopes from four libraries:CENDL-3.2,JEFF-3.3,JENDL-5,and ENDF/B-Ⅷ.0 were used *** comparing the simulation and experimental results,improvements and deficiencies in the evaluated nuclear data of the Pb isotopes were *** of the calculated results were consistent with the experimental results;however,a few areas did not fit *** the(n,el)energy range,the simulated results from CENDL-3.2 were significantly overestimated;in the(n,inl)D and the(n,inl)C energy regions,the results from CENDL-3.2 and ENDF/B-Ⅷ.0 were significantly overestimated at 120°,and the results from JENDL-5 and JEFF-3.3 are underestimated at 60°in the(n,inl)D energy *** calculated spectra were analyzed by comparing them with the experimental spectra in terms of the neutron spectrum shape and C/E *** results indicate that the theoretical simulations,using different data libraries,overestimated or underestimated the measured values in certain energy *** neutron energies and angular distributions in

关键词： Integral experiment Neutron leakage spectra ^(nat)Pb D-T and D-D neutron sources Evaluated nuclear data

来源：评论

学校读者我要写书评

暂无评论

Acceleration by Stepsize Hedging: Multi-Step Descent and the Silver Stepsize Schedule

引用

Journal of the ACM 2025年第2期72卷 1-38页

作者： Altschuler, Jason M. Parrilo, Pablo A. Statistics and Data Science University of Pennsylvania Philadelphia United States EECS Massachusetts Institute of Technology Cambridge United States

Can we accelerate the convergence of gradient descent without changing the algorithm - just by judiciously choosing stepsizes? Surprisingly, we show that the answer is yes. Our proposed Silver Stepsize Schedule optimizes strongly convex functions in iterations, where is the silver ratio and κ is the condition number. This is intermediate between the textbook unaccelerated rate κ and the accelerated rate due to Nesterov in 1983. The non-strongly convex setting is conceptually identical, and standard black-box reductions imply an analogous partially accelerated rate . We conjecture and provide partial evidence that these rates are optimal among all stepsize *** Silver Stepsize Schedule is constructed recursively in a fully explicit way. It is non-monotonic, fractal-like, and approximately periodic of period . This leads to a phase transition in the convergence rate: initially super-exponential (acceleration regime), then exponential (saturation regime).The core algorithmic intuition is hedging between individually suboptimal strategies - short steps and long steps - since bad cases for the former are good cases for the latter, and vice versa. Properly combining these stepsizes yields faster convergence due to the misalignment of worst-case functions. The key challenge in proving this speedup is enforcing long-range consistency conditions along the algorithm's trajectory. We do this by developing a technique that recursively glues constraints from different portions of the trajectory, thus removing a key stumbling block in previous analyses of optimization algorithms. More broadly, we believe that the concepts of hedging and multi-step descent have the potential to be powerful algorithmic paradigms in a variety of contexts in optimization and *** article publishes and extends the first author's 2018 master's thesis (advised by the second author) - which established for the first time that judiciously choosing stepsizes can enable acceleration in conve

关键词： Convex optimization

来源：评论

学校读者我要写书评

暂无评论

An electron-transfer-tuning strategy at the graphene/metal interface for improving acidic water electrolysis in proton exchange membrane electrolyzers

引用

Journal of Energy Chemistry 2025年第4期103卷 344-352页

作者： Yue Xu Shuaidong Li Yingjian He Fumiya Shiokawa Samuel Jeong Aimi Asilah Haji Tajuddin Zeyu Xi Yoshikazu Ito Jingzi Zhang Zeyun Cai Xi Lin Kailong Hu School of Materials Science and Engineering and Institute of Materials Genome&Big DataHarbin Institute of TechnologyShenzhen 518055GuangdongChina Institute of Special Environments Physical Sciences Harbin Institute of TechnologyShenzhen 518055GuangdongChina School of Computer Science and Technology Harbin Institute of TechnologyShenzhen 518055GuangdongChina Institute of Applied Physics Graduate School of Pure and Applied SciencesUniversity of TsukubaTsukuba 305-8573Japan

Graphene encapsulation has been shown to be an effective technique for improving the corrosion resistance of non-noble metal catalysts for the acidic water *** key challenge lies in enhancing the electrocatalytic activity of graphene-encapsulated metals while maintaining their durability in acidic ***,an electron-transfer-tuning strategy is investigated at the graphene/NiMo interface,aiming to improve the hydrogen evolution reaction(HER)performance of graphene-encapsulated NiMo *** doping of Ti,a low electronegativity element,into NiMo substrate was confirmed to increase electron transfer from the metal core toward the *** electron-rich state on graphene facilitates the adsorption of positively charged protons on graphene,thereby enabling a Pt/C-comparable performance in 0.5 M H_(2)SO_(4),with only a 3.8%degradation in performance over a 120-h continuous *** proton exchange membrane(PEM)water electrolyzer assembled by the N-doped grapheneencapsulated Ti-doped NiMo exhibits a smaller cell voltage to achieve a current density of 2.0 A cm^(-2),in comparison to the Pt/C based *** study proposes a novel electron-transfer-tuning strategy to improve the HER activity of graphene-encapsulated non-noble metal catalysts without sacrificing durability in acidic electrolytes.

关键词： Hydrogen evolution Acidic media Graphene encapsulation Interface Proton exchange membrane

来源：评论

学校读者我要写书评

暂无评论

FedLRes: enhancing lung cancer detection using federated learning with convolution neural network (ResNet50)

引用

Neural Computing and Applications 2025年第14期37卷 8273-8284页

作者： Usharani, C. Selvapandian, A. Department of Artificial Intelligence and Data Science Ramco Institute of Technology Rajapalayam India Department of Electronics and Communication Engineering Gnanamani College of Technology Namakkal India

Lung cancer remains a leading global cause of mortality, necessitating efficient early detection. Lung cancer image analysis plays a pivotal role, yet current manual segmentation by oncologists is laborious. Our innovative (FedLRes) approach proposes a comprehensive system for automated diagnoses of lung cancer using federated learning with ResNet50. The lung cancer dataset from the Iraq-Oncology Teaching Hospital/National Centre for Cancer Diseases was gathered by the IQ-OTH/NCCD. The dataset consists of 1097 images, split into training (822 images) and validation (275 images) sets. The training set includes 312 normal, 420 malignant, and 90 benign cases. The training set's data are further enhanced by using data augmentation techniques, which are then applied to normalise images differences before being fed into the federated learning with ResNet50 architecture. This approach combines deep learning models trained on different datasets allowing for improvising accuracy and generalisation. The federated learning approach enables the use of distributed data while ensuring data privacy and security. The proposed approach compared with different state of art algorithms. Through rigorous experimentation, our system showcases remarkable advancements a classification accuracy of 99.40%. This innovative approach, utilising 3D input CT scan data, offers a potent and precise tool for early detection and effective treatment strategies against the scourge of lung cancer. © The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature 2025.

关键词： Lung cancer

来源：评论

学校读者我要写书评

暂无评论

StealthFace:Transfer Learning-Based Ensemble Model for Disguised Face Recognition using Skin-Segmented Images

IAENG International Journal of Computer Science

引用

IAENG International Journal of Computer science 2025年第4期52卷 1244-1260页

作者： Padmashree, G. Kotegar, Karunakar A Department of Data Science and Computer Applications Manipal Institute of Technology Manipal Academy of Higher Education Karnataka Manipal576104 India

Disguised face identification is challenging since people cover their identities by wearing masks, hats, sunglasses, or other disguises. These disguises dramatically modify face features, making identifying individuals a difficult task. In this study, we introduce StealthFace, an ensemble model that utilizes transfer learning and skin-segmented images to improve the recognition of disguised faces. To improve the accuracy and resilience of disguised face recognition, StealthFace leverages the power of deep learning algorithms, transfer learning, and ensemble methods. The framework employs pre-trained convolutional neural network models, such as ResNet50 and DenseNet121 for feature extraction and to learn discriminative features that are important in identifying individuals despite the disguises. The ensemble approach combines models’ predictions, utilizing their collective expertise and capturing multiple perspectives on disguised faces. Our ensemble model outperforms other state-of-the-art methods with an overall accuracy of 99.24% and contributes to enhancing security measures and advancing the development of face recognition systems. © (2025), (International Association of Engineers). All Rights Reserved.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

Understanding the Aftermath of COVID-19: A data-Driven Exploration of Health Effects 3

Understanding the Aftermath of COVID-19: A Data-Driven Explo...

引用

3rd International Conference on Intelligent data Communication Technologies and Internet of Things, IDCIoT 2025

作者： Selvaraj, Arikumar K. Naveen Akash, P. Venkata Swamy, B. Department of Data Science and Business Systems SRM Institute of Science and Technology Kattankulathur India

ISBN: (纸本)9798331527549

The COVID-19 pandemic has ravaged public health in a manner that is unprecedented: millions of infections and fatalities worldwide. Now that the immediate crisis seems to recede, there is growing concern over long-term health consequences faced by survivors, commonly referred to as "long COVID." This study aims to systematically explore the multifaceted health effects experienced by individuals who have recovered from COVID-19, focusing on physical, mental, and cognitive health issues that have emerged in the aftermath of the infection. Employing a data-driven approach, the collected extensive survey data from a diverse cohort of COVID-19 survivors to assess their post-recovery health status. The survey included questions targeting various health dimensions, such as fatigue, pain, anxiety, depression, and cognitive function. To analyze the data, utilized several machine learning algorithms to identify significant health complications and to determine the interrelationships among various health factors. © 2025 IEEE.

关键词： COVID-19

来源：评论

学校读者我要写书评

暂无评论

H2O Guard: Predicting Water Potability with Machine Learning and Blockchain Integration 3

H2O Guard: Predicting Water Potability with Machine Learning...

引用

3rd International Conference on Electronics and Renewable Systems, ICEARS 2025

作者： Nair, Rahul Aalim, Ansab Manhaas, Ishaan Rastogi, Kshitij Kalpana, A.V. Srm Institute of Science and Technology Department of Data Science and Business Systems Kattankulathur India

ISBN: (纸本)9798331509675

Access to clean drinking water is critical to public health and well - being. Water compatibility refers to the safety of drinking water for human consumption and includes various physical, chemical and biological parameters. Contaminated water can cause serious health problems, particularly among vulnerable populations due to water-borne diseases. Traditionally, water quality assessment relies on time - consuming laboratory tests that can delay responses to contamination. The H2OGuard project presents a novel approach: It uses deep learning and machine learning algorithms to predict water sustainability based on key environmental factors such as pH, turbidity and chemical contaminants. With these advanced methods, we aim to improve the efficiency and accuracy of water quality testing. In addition, the integration of blockchain technology is intended to ensure secure and immutable storage of water quality data and promote integrity and transparency. This paper discusses the methods of the H2OGuard project, evaluates the performance of various predictive models, and highlights the critical importance of blockchain in securing data. With this innovative framework, we aim to enable more efficient and timely assessment of drinking water safety, ultimately contributing to improved public health outcomes. © 2025 IEEE.

关键词： Electronic health record

来源：评论

学校读者我要写书评

暂无评论

Enhancing Explainability of Deep Learning-Based ECG Diagnosis Using Large Language Models 24

Enhancing Explainability of Deep Learning-Based ECG Diagnosi...

引用

8th International Conference on Advances in Artificial Intelligence, ICAAI 2024

作者： Wu, Shi Zhou, Jianlong Dong, Yifei Chen, Fang Data Science Institute University of Technology Sydney SydneyNSW Australia

ISBN: (纸本)9798400718014

The electrocardiogram (ECG) is an essential diagnostic tool for monitoring heart health. Traditional manual methods for ECG interpretation are increasingly challenged by the complexity of heart diseases and the volume of ECG data. Recent advancements in artificial intelligence (AI), particularly deep learning, have improved the efficiency and accuracy of ECG diagnostics. However, the "black-box" nature of AI models poses trust and verification challenges. To address this, we propose a novel approach integrating Explainable AI (XAI) techniques with multimodal large language models to enhance ECG diagnostic interpretability. Our methodology employs a 2D convolutional neural network (CNN) to classify ECG signals, followed by Grad-CAM to generate heatmaps highlighting critical areas influencing AI decisions. These enhanced ECG images and their classifications are then input into a multimodal large language model to produce comprehensive explanatory outputs. This approach combines the visual processing power of CNNs with the contextual understanding of multimodal models, offering clearer insights into AI reasoning. Our research demonstrates improved diagnostic accuracy and interpretability, setting a new standard for AI integration in clinical practices. Specifically, our method achieved an F1-score of 0.67 for ECG classification using Inception and a BERT-Score of 0.818 for text generation using Gemini_GradCAM. © 2024 Copyright held by the owner/author(s).

关键词： Convolutional neural networks

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：