检索结果-内蒙古大学图书馆

Densification behavior of ultrafine WC-based composites with AlxCoCrCuFeNi high-entropy alloy binders

INTERNATIONAL JOURNAL OF APPLIED CERAMIC TECHNOLOGY 2025年第3期22卷

作者： Luo, Wenyan Liu, Yunzhong Hubei Polytech Univ Sch Mech & Elect Engn Huangshi Peoples R China South China Univ Technol Natl Engn Res Ctr Near Net Shape Forming Met Mat Guangzhou 510640 Peoples R China Hubei Polytech Univ Hubei Key Lab Intelligent Convey Technol & Device Huangshi Peoples R China

Ultrafine WC-based cemented carbides with 10 wt.%AlxCoCrCuFeNi high-entropy alloy (HEA) binders were fabricated by spark plasma sintering, and the effects of HEA binders on the densification behavior of the WC-HEA cemented carbides were studied. The densification of the WC-HEA cemented carbide, as well as traditional WC-Co, can be divided into the slow densification stage, rapid densification stage, and final densification stage. The densification behavior of the WC-HEA cemented carbides largely depends on the performance of the HEA binder during the SPS process. The sluggish diffusion effect of HEA weakens the diffusion of W atom on the powder particle surface and inhibits the growth of WC grain. Consequently, the disappearance of pores in the sintered compact is hindered, which leads to a low relative density of the WC-HEA cemented carbide. With the increase of Al content, the inhibitory effect of the AlxCoCrCuFeNi binder on the growth of WC grain is suppressed, and an Al-containing phase with a low melting point is more likely to form. Therefore, the relative density of the WC-10 wt.%AlxCoCrCuFeNi cemented carbide raises linearly with increasing Al content of the HEA binder.

关键词： densification high-entropy alloy binder spark plasma sintering ultrafine grain WC-based cemented carbide

来源：评论

学校读者我要写书评

暂无评论

An Adaptive Continual Learning Method for Nonstationary Industrial Time Series Prediction

引用

IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS 2025年第2期21卷 1160-1169页

作者： Wu, Mengqing Zhou, Xiaofeng Li, Shuai Shi, Haibo Chinese Acad Sci Key Lab Networked Control Syst Shenyang 110016 Peoples R China Chinese Acad Sci Shenyang Inst Automat Shenyang 110016 Peoples R China Chinese Acad Sci Inst Robot & Intelligent Mfg Shenyang 110169 Peoples R China Univ Chinese Acad Sci Beijing 100049 Peoples R China

Deep learning models have gained significant attention and application in recent years to improve the accuracy and efficiency of industrial time series prediction. However, the dynamic changes in industrial processes present a key challenge for data-driven models. Specifically, the performance of deployed models deteriorates over time and fails to adapt to new operating conditions. Currently, two common update methods exist: Retraining the model using historical and new operating data, which incurs high computation and storage costs, or incrementally fine-tuning the model solely using new data, which leads to catastrophic forgetting of learned patterns. To address these issues, this article proposes an adaptive continual learning method for nonstationary industrial time series prediction. Our approach tackles the problems by hint-based network parameter learning to retain the dark knowledge from previous tasks and avoid catastrophic forgetting of accumulated knowledge. In addition, we design a soft buffer to aid memory and learning of key patterns under the current operating condition. Lastly, a time-sensitive activation function is proposed to endow the neural network with time-evolving properties, thereby enhancing the model's generalization ability. Compared with other update methods and different continual learning methods, the superiority of our method is validated on solar power generation data and real data of grinding and grading process.

关键词： Continuing education Adaptation models Predictive models Data models Training Time series analysis Computational modeling Production Optimization Service robots Adaptive continual learning catastrophic forgetting generalization nonstationary industrial process time series prediction

来源：评论

学校读者我要写书评

暂无评论

Effect of different imidazole moieties on luminescence properties of 3,8,13-trisubstituted triazatruxene derivatives

引用

JOURNAL OF MOLECULAR STRUCTURE 2025年 1329卷

作者： La, Weibo Yu, Tianzhi Zhao, Yuling Zhang, Di Tian, Qingyuan Yang, Bosheng Su, Wenming Lanzhou Jiaotong Univ Minist Educ Key Lab Optoelect Technol & Intelligent Control Lanzhou 730070 Peoples R China Lanzhou Jiaotong Univ Sch Chem & Chem Engn Lanzhou 730070 Peoples R China Chinese Acad Sci Suzhou Inst Nanotech & Nanobion Printable Elect Res Ctr Suzhou 215123 Peoples R China

Two new 3,8,13-trisubstituted triazatruxene derivatives containing different imidazole-derived moieties, 3,8,13TPI-TAT and 3,8,13-TTPI-TAT, were successfully synthesized by grafting different imidazole-derived moieties (phenanthroimidazole (PI) and 1,4,5-triphenylimidazole (TPI)) onto the 3,8,13-positions of a triazatruxene (TAT) core, and characterized by NMR (1H and 13C), high resolution mass spectrometry. Using the already reported 3,8,13-TBI-TAT as a comparison, the thermal stability, photophysical and electrochemical properties of these compounds were comparatively investigated to reveal the effects of different imidazole-derived groups on the luminescence properties of the compounds. The doped devices with a structure of ITO/PEDOT:PSS (45 nm)/ PVK:PBD (7:3, wt%):TAT compounds (x wt%) (30 nm)/TPBi (20 nm)/Liq (2 nm)/Al (150 nm) and non-doped devices were fabricated by solution-processed the emitting layers to investigate their electroluminescence (EL) performances, in which the devices of the diphenylimidazole substituted compound (3,8,13-TDPI-TAT) showed the best EL performances compared with that of the other two compounds. The doped devices of 3,8,13-TTPITAT exhibited a maximum luminance (Lmax) of 402 cd/m2 and a maximum external quantum efficiency (EQEmax) of 0.72 %, while its non-doped devices showed Lmax of 500 cd/m2 and the EQEmax of 1.43 %. It has important potential significance for the further design and synthesis of TAT derivatives with good charge transport and high EQE.

关键词： Triazatruxenederivative Phenanthro[d]imdazole Benzo[d]-imidazole 5-triphenylidazole Photoluminescence Electroluminescence

来源：评论

学校读者我要写书评

暂无评论

MLDA-Net: Multi-Level Deep Aggregation Network for 3D Nuclei Instance Segmentation

引用

IEEE JOURNAL OF BIOMEDICAL AND HEALTH INFORMATICS 2025年第5期29卷 3516-3525页

作者： Hu, Bin Ye, Zhiwei Wei, Zimei Snezhko, Eduard Kovalev, Vassili Ye, Mang Hubei Univ Technol Sch Comp Wuhan 430068 Peoples R China Hubei Prov Key Lab Green Intelligent Comp Power Ne Wuhan 430068 Peoples R China Natl Acad Sci Belarus United Inst Informat Problems Minsk 220012 BELARUS Wuhan Univ Sch Comp Sci Wuhan 430072 Peoples R China

Segmentation of cell nuclei from three-dimensional (3D) volumetric fluorescence microscopy images is crucial for biological and clinical analyses. In recent years, convolutional neural networks have become the reliable 3D medical image segmentation standard. However, convolutional layers are limited by their finite receptive fields and weight-sharing mechanisms. Consequently, they struggle to effectively model long-range dependencies and spatial correlations, which may lead to inadequate nuclei segmentation. Moreover, the diversity in nuclear appearance and density poses additional challenges. This work proposes a lightweight multi-layer deep aggregation network, MLDA-Net, incorporating Wide Receptive Field Attention (WRFA). This module effectively simulates the large receptive field generated by self-attention in the Swin Transformer while requiring fewer model parameters. This design implements an extended global sensory field that enhances the ability to capture a wide range of spatial information. In addition, the multiple cross-attention (MCA) module in MLDA-Net enhances the output features of different resolutions from the encoder while maintaining global effectiveness. The Multi-Path Aggregation Feature Pyramid Network (MAFPN) receives multi-scale outputs from the MCA module, generating a robust hierarchical feature pyramid for the final prediction. MLDA-Net outperforms state-of-the-art networks, including 3DU-Net, nnFormer, UNETR, SwinUNETR, and 3DUXNET, on the 3D volumetric datasets NucMM and MitoEM. It achieves average performance improvements of 4% to 7% in F1 score, MIoU, and PQ metrics, thereby establishing new benchmark results.

关键词： Three-dimensional displays Image segmentation Feature extraction Transformers Computer architecture Instance segmentation Bioinformatics Microprocessors Convolution Kernel 3D instance segmentation nuclei segmentation feature aggregation large receptive field deep learning

来源：评论

学校读者我要写书评

暂无评论

Sequential Spiking Neural P Systems with Polarizations Based on Minimum Spike Number Working in the Accepting Mode 16th

Sequential Spiking Neural P Systems with Polarizations Base...

引用

16th International Conference on Bio-Inspired computing: Theories and Applications, BIC-TA 2021

作者： Liu, Li Jiang, Keqin The University Key Laboratory of Intelligent Perception and Computing of Anhui Province Anqing Normal University Anhui Anqing246133 China School of Computer and Information Anqing Normal University Anhui Anqing246133 China

ISBN: (纸本)9789811912559

Inspired by the information transmission method of electrical signals in the biological impulse nervous system, a new variant of the spiking neural P systems, called spiking neural P systems with polarizations (PSN P systems for short), is proposed. In this research, the excitation conditions of PSN P systems are mainly determined by the polarization and the number of spikes together. Based on the fact that spiking neural P systems can be used as a different working device, the computational power of the sequential spiking neural P systems with polarization induced by the number of spikes can work in the receptive mode. Specifying the computational result as the temporal distance between the first two spike moments of input neuron reception proves that the PSN P systems are Turing universal as number accepting devices. © 2022, Springer Nature Singapore Pte Ltd.

关键词： Polarization

来源：评论

学校读者我要写书评

暂无评论

Research on Quantitative Safety Risk Assessment Methods at Vehicle Controllability Level

引用

IEEE Access 2025年 13卷 82527-82539页

作者： He, Lei Li, Zichong Yang, Xuezhu Zhou, Yuanle Qu, Zikun Ren, Zhiqiang Jilin University State Key Laboratory of Automotive Simulation and Control Jilin Changchun130025 China China First Automobile Works Group Corporation Ltd. Jilin Changchun130000 China

In recent years, advancements in electrification, intelligent connectivity, and autonomous driving have made functional safety, particularly the vehicle-level controllability, a critical research focus. Although the ISO 26262 standard provides a systematic framework for automotive functional safety, challenges remain in accurately defining and quantifying controllability levels. In this paper, we analyze the influence of potential faults—such as unintended torque and loss of regenerative braking force—on vehicle controllability under various operating conditions. A database of vehicle attitude data and corresponding controllability levels is established through real-vehicle tests. Feature screening is then performed to identify the most relevant attributes that significantly affect controllability. Based on this database, a controllability model is developed using the Support Vector machine (SVM) method, which can effectively predict controllability levels. The model is applied in multiple simulation scenarios to assess vehicle controllability under different conditions. Additionally, the safety boundaries of motor torque are determined at varying speeds, providing insights into critical thresholds beyond which the vehicle’s controllability may be compromised. This work presents a novel approach for the quantitative analysis of vehicle controllability and safety risks, offering a reliable tool for evaluating functional safety in modern vehicles. © 2013 IEEE.

关键词： ISO Standards

来源：评论

学校读者我要写书评

暂无评论

Enhancing knowledge tracing with fine-grained session modeling

引用

INTERNATIONAL JOURNAL OF machine LEARNING AND CYBERNETICS 2025年 1-17页

作者： Wang, Jing Ma, Huifang Zhang, Mengyuan Li, Zhixin Chang, Liang Northwest Normal Univ Coll Comp Sci & Engn Lanzhou 730070 Gansu Peoples R China Guangxi Normal Univ Key Lab Educ Blockchain & Intelligent Technol Minist Educ Guilin 541004 Guangxi Peoples R China Guilin Univ Elect Technol Sch Comp Sci & lnformat Secur Guilin 541004 Guangxi Peoples R China

Knowledge tracing (KT) aims to dynamically model learners' evolving knowledge states based on their historical learning records, playing a vital role in online education systems. Most existing KT methods learn the knowledge states as a transition pattern from the previous exercise to the next one, treating learners' entire learning records as continuous and uniformly distributed. However, we argue that actual learning records can be divided into distinct shorter sessions. To this end, we propose a novel KT model called Fine-grained Session Modeling for Knowledge Tracing (FSM4KT), which is designed to capture learners' knowledge state changes with finer granularity. In particular, we first divide learners' extensive historical learning records into shorter sessions from either temporal or knowledge concept-related perspective. Subsequently, a dedicated designed session-based knowledge proficiency modeling component is presented, which figures out intra-session and inter-session fine-grained interaction dependencies and knowledge state changes. Moreover, a global knowledge proficiency modeling component is introduced to holistically model learners' knowledge states. Extensive experimental results on three real-world datasets demonstrate that FSM4KT outperforms most of the current baseline methods, thus proving the effectiveness of FSM4KT.

关键词： intelligent education Knowledge tracing Correctness prediction Graph neural networks

来源：评论

学校读者我要写书评

暂无评论

Multiple adverse weather image restoration: A review

引用

NEUROcomputing 2025年 618卷

作者： Xiao, Hanguang Liu, Shihong Zuo, Kun Xu, Haipeng Cai, Yuyang Liu, Tianqi Yang, Zhiying Chongqing Univ Technol Sch Artificial Intelligence Chongqing 401135 Peoples R China Univ Shanghai Sci & Technol Shanghai 200093 Peoples R China Chongqing Key Lab Met Intelligent Equipment Chongqing 401122 Peoples R China CISDI Res & Dev Co Ltd Chongqing 401122 Peoples R China

Adverse weather conditions significantly impact the machine vision perception of unmanned platforms such as drones and autonomous vehicles. Therefore, restoring and enhancing images affected by these conditions is crucial. However, there is a lack of a comprehensive systematic review on the latest research progress in unified weather image restoration under all-in-one weather conditions and other factors. In this paper, we survey recent progress in various network architectures for multitask image restoration. Specifically, we review and compare different methods including, U-Net, GAN, Transformer, and U-Net/Transformer hybrids. Additionally, we evaluate publicly available datasets for single-task scenarios and compare the performance of these methods comprehensively, and analyze corresponding evaluation metrics. Based on these findings, we believe that Transformer and U-Net models are particularly promising for multi-task image restoration. Nevertheless, further research is needed to fully explore this area. Researchers can improve the image restoration effect from aspects such as data dynamic flow, encoder-decoder internal structure, etc. These insights contribute to advancing image acquisition technologies and addressing challenges in military image information retrieval, including military reconnaissance, terrain analysis, target identification, and battlefield surveillance.

关键词： Image restoration Deep learning All-In-One Adverse weather Transformer

来源：评论

学校读者我要写书评

暂无评论

Salix-Leaf: Find Main Veins of Signal Clusters for Practical Parallel Decoding

引用

IEEE Transactions on Mobile computing 2025年

作者： Li, Yajun Zhao, Jumin Li, Dengao Wu, Hejun Xu, Shuang Bai, Ruiqin Taiyuan University of Technology College of Electronic Information and Optical Engineering Taiyuan China Key Laboratory of Big Data Fusion Analysis and Application of Shanxi Province Taiyuan China Intelligent Perception Engineering Technology Center of Shanxi Taiyuan China Taiyuan University of Technology College of Computer Science and Technology College of Data Science Taiyuan China Sun Yat-sen University Department of Computer Science Guangzhou China

Parallel decoding of backscatter improves communication throughput by enabling concurrent transmission of backscatter tags. In practical applications of parallel decoding, it is extremely difficult to distinguish collided signals in superclusters where multiple signal clusters overlap. Existing methods are usually effective for superclusters with uniformly distributed signals. Nevertheless, there are many more scenarios in which signals in superclusters tend to gather unevenly, and existing methods cannot work. Such uneven clustering of signals occurs due to the following two possible causes: (1) signal-strengthdifferences (SSDs) among tags;or (2) cluster drifting (CD) driven by interferences from other objects within communication environments. This paper proposes a novel scheme called SalixLeaf, which aims to identify the main veins of signal clusters to address this problem of superclusters with unevenly distributed signals. Salix-Leaf identifies the main vein of each signal cluster for fine-grained clustering so that the direction of the main veins can be used to verify the accuracy of clustering. In addition, SalixLeaf employs a supercluster decomposer that divides signals into different segments for clustering analysis, enhancing robustness and practicability. Experimental results show that Salix-Leaf achieves a 1.2-fold increase in throughput and a 25% reduction in bit error rate (BER) compared to the state-of-the-art. © 2002-2012 IEEE.

关键词： Backscattering

来源：评论

学校读者我要写书评

暂无评论

Stability and convergence rate analysis of continuous-time delay-difference systems with both point and distributed delays

引用

INTERNATIONAL JOURNAL OF SYSTEMS SCIENCE 2025年第3期56卷 434-449页

作者： Zhang, Qianqian Li, Zhao-Yan Zhou, Bin Harbin Inst Technol Sch Math Harbin 150001 Peoples R China Civil Aviat Flight Univ China Sch Sci Chengdu Peoples R China Harbin Inst Technol Ctr Control Theory & Guidance Technol Harbin Peoples R China Natl Key Lab Complex Syst Control & Intelligent Ag Harbin Peoples R China

This paper studies the exponential stability and convergence rate analysis of continuous-time delay-difference systems. Firstly, stability and convergence rate analysis of delay-difference systems with both point delays and distributed delays having exponential integral kernels are studied by using the weighted Lyapunov-Krasovskii functionals (LKFs) approach and the state transformation approach, respectively. Different sufficient stability conditions expressed by linear matrix inequalities (LMIs) are presented. Secondly, as a particular case, LMIs based conditions and spectral radius based conditions are given to ensure the exponential stability with a guaranteed convergence rate for delay-difference systems with both point delays and distributed delays having constant integral kernels, respectively. Finally, numerical examples illustrate the effectiveness of the obtained results.

关键词： Delay-difference systems exponential stability convergence rate analysis Lyapunov-Krasovskii functionals

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：