检索结果-内蒙古大学图书馆

Research on accurate virtual trajectory length model for TGS transmission measurement

nuclear Science and Techniques 2025年第3期36卷 178-188页

作者： Rong-Rong Su San-Gang Li Chu-Xiang Zhao Li Yang Ming-Zhe Liu Shan Liao Zhi Zhou Qing-Shan Tan Zhi-Xing Gu Xian-Guo Tuo Yi Cheng College of Nuclear Technology and Automation Engineering Chengdu University of Technology1#DongsanluErxianqiaoChengdu 610059China Applied Nuclear Technology in Geosciences Key Laboratory of Sichuan Province (Chengdu University of Technology)1#DongsanluErxianqiaoChengdu 610059China University of Science and Technology of China Hefei 230026China School of Physics and Electronic Engineering Sichuan University of Science and EngineeringZigong 643000China

To accurately reconstruct the tomographic gamma scanning(TGS)transmission measurement image,this study optimized the transmission reconstruction equation based on the actual situation of TGS transmission *** the transmission reconstruction equation and the Monte Carlo program Geant4,an innovative virtual trajectory length model was *** model integrated the solving process for the trajectory length and detection efficiency within the same *** mitigate the influence of the angular distribution ofγ-rays emitted by the transmitted source at the detector,the transport processes of numerous particles traversing a virtual nuclear waste barrel with a density of zero were ***,a certain amount of information was captured at each step of particle ***,the model addressed the nonuniform detection efficiency of the detector end face by considering whether the energy deposition of particles in the detector equaled their initial *** models were established to validate the accuracy and reliability of the virtual trajectory length *** 1 was a simplified nuclear waste barrel,whereas Model 2 closely resembled the actual structure of a nuclear waste *** results indicated that the proposed virtual trajectory length model significantly enhanced the precision of the trajectory length determination,substantially increasing the quality of the reconstructed *** example,the reconstructed images of Model 2 using the“point-to-point”and average trajectory models revealed a signalto-noise ratio increase of 375.0%and 112.7%,***,the virtual trajectory length model proposed in this study holds paramount significance for the precise reconstruction of transmission ***,it can provide support for the accurate detection of radioactive activity in nuclear waste barrels.

关键词： Tomographic gamma scanning Transmission measurement reconstruction Geant4 Trajectory length model Nonuniform detection efficiency

来源：评论

学校读者我要写书评

暂无评论

Power Allocation Optimization of Hybrid Energy Storage System Based on AOA-VMD 11th

Power Allocation Optimization of Hybrid Energy Storage Syste...

引用

11th Frontier Academic Forum of Electrical engineering, FAFEE 2024

作者： Liu, Wei Zhao, Wenhao Feng, Chengwu Jing, Hailian Fang, Xiao Liu, Yingcan Han, Yuecheng College of Nuclear Technology and Automation Engineering Chengdu University of Technology Chengdu610059 China

ISBN: (纸本)9789819787791

With the construction and grid integration of large-scale photovoltaic power generation systems, utilizing energy storage technology to reduce grid-connected power fluctuations and enhance grid stability has become a research hotspot. This paper, based on a hybrid energy storage system composed of flywheels and lithium-ion batteries, analyzes the measured photovoltaic output power, establishes a hybrid energy storage system model to smooth the fluctuation rate of photovoltaic power generation. Addressing the power allocation issue of the hybrid energy storage system, an optimization algorithm (Arithmetic Optimization Algorithm, AOA) combined with Variational Mode Decomposition (VMD) is employed to solve the model. The simulation results demonstrate that the AOA-VMD algorithm reduces the rated power of both lithium batteries and flywheels. This validates its superiority in smoothing photovoltaic output fluctuations and addressing the mode mixing problem of non-stationary signals, as well as its effectiveness in optimizing power allocation in hybrid energy storage systems. It enhances the efficiency and operational reliability of hybrid energy storage systems, thus advancing the large-scale application of renewable energy sources. © Beijing Paike Culture Commu. Co., Ltd. 2025.

关键词： Hybrid power

来源：评论

学校读者我要写书评

暂无评论

A nuclide identification method ofγspectrum and model building based on the transformer

引用

nuclear Science and Techniques 2025年第1期36卷 80-93页

作者： Fei Li Chu-Yang Luo Ying-Zi Wen Sheng Lv Feng Cheng Guo-Qiang Zeng Jian-Feng Jiang Bing-Hai Li College of Nuclear Technology and Automation Engineering Chengdu University of TechnologyChengdu 610000China Applied Nuclear Technology in Geosciences Key Laboratory of Sichuan Province Chengdu 610000China Department of Physics Universitat Politècnica de CatalunyaBarcelona 08028Spain Airborne Survey and Remote Sensing Center of Nuclear Industry Shijiazhuang 050000China Hebei Key Laboratory of Airborne Detection and Remote Sensing Technology Shijiazhuang 050000China

In current neural network algorithms for nuclide identification in high-background,poor-resolution detectors,traditional network paradigms including back-propagation networks,convolutional neural networks,recurrent neural networks,etc.,have been limited in research on γ spectrum analysis because of their inherent mathematical *** is difficult to make progress in terms of training data requirements and prediction *** contrast to traditional network paradigms,network models based on the transformer structure have the characteristics of parallel computing,position encoding,and deep stacking,which have enabled good performance in natural language processing tasks in recent ***,in this paper,a transformer-based neural network (TBNN) model is proposed to achieve nuclide identification for the first ***,the Geant4 program was used to generate the basic single-nuclide energy spectrum through Monte Carlo simulations.A multi-nuclide energy spectrum database was established for neural network training using random matrices of γ-ray energy,activity,and *** on the encoder–decoder structure,a network topology based on the transformer was built,transforming the 1024-channel energy spectrum data into a 32×32 energy spectrum sequence as the model *** experiments and adjustments of model parameters,including the learning rate of the TBNN model,number of attention heads,and number of network stacking layers,the overall recognition rate reached 98.7%.Additionally,this database was used for training AI models such as back-propagation networks,convolutional neural networks,residual networks,and long shortterm memory neural networks,with overall recognition rates of 92.8%,95.3%,96.3%,and 96.6%,*** indicates that the TBNN model exhibited better nuclide identification among these AI models,providing an important reference and theoretical basis for the practical application of transformers in the qualitative and quantitativ

关键词： Nuclide identification Neural network Transformer

来源：评论

学校读者我要写书评

暂无评论

Enhancing traffic light recognition in autonomous vehicles: A novel convolutional neural network approach for fog removal using efficient channel attention 4

Enhancing traffic light recognition in autonomous vehicles: ...

引用

4th International Conference on Computer Vision, Application, and Algorithm, CVAA 2024

作者： Ji, Shuai Zhang, Huangnian Dong, Xingwen The College of Nuclear Technology and Automation Engineering Chengdu University of Technology Sichuan Chengdu610059 China

ISBN: (数字)9781510687622

ISBN: (纸本)9781510687615

Adverse environmental conditions such as haze, rain, and dust can significantly impair visibility for vehicles in motion, leading to blurred vision. Traditional de-fogging techniques often encounter limitations, including reliance on pre-existing assumptions, color distortion, and inadequate real-time performance. Consequently, these methods may fail to effectively distinguish traffic light images post-de-fogging, thereby hindering target recognition in autonomous driving systems. This study introduces a convolutional neural network that integrates deep and shallow feature fusion utilizing the Efficient Channel Attention mechanism. By employing an atmospheric scattering model, shallow features from fogged images are extracted through convolutional layer operations, while deep features are obtained using parallel multi-scale convolution kernels. The integration of shallow and deep features is achieved through a series of fusion steps, enhanced by the application of an effective channel attention mechanism. Ultimately, a transmittance map corresponding to the fog map is derived through nonlinear regression, facilitating the recovery of fog-free images and improving the recognition capabilities of the YOLOv8 model. Experimental results demonstrate that this approach effectively identifies traffic light targets during simulations of vehicles navigating traffic intersections. Furthermore, the quality of fog removal achieved by this method surpasses that of other algorithms, as evidenced by both subjective evaluations and objective analyses. © 2025 SPIE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Masked hybrid attention with Laplacian query fusion and tripartite sequence matching for medical image segmentation

引用

Neural Computing and Applications 2025年第8期37卷 5891-5911页

作者： Ekong, Favour Yu, Yongbin Patamia, Rutherford Agbeshi Sarpong, Kwabena Ukwuoma, Chiagoziem C. Wang, Xiangxiang Ukot, Akpanika Robert Cai, Jingye School of Information and Software Engineering University of Electronic Science and Technology of China Sichuan Chengdu611731 China College of Nuclear Technology and Automation Engineering Chengdu University of Technology Sichuan Chengdu610059 China Sichuan Engineering Technology Research Center for Industrial Internet Intelligent Monitoring and Application Chengdu University of Technology Sichuan Chengdu610059 China

Medical image segmentation is pivotal in computer-aided diagnosis systems, demanding high precision and contextual understanding. Vision Transformer-based approaches have gained much attention recently due to their excellent performance and ability to capture long-range dependencies in medical images. However, research shows they suffer from inadequate multi-scale feature integration, poor object localization, and inconsistent mask predictions, leading to sub-optimal segmentation performance. This paper addresses these challenges by redefining semantic medical image segmentation through learnable object queries within an enhanced transformer framework with a masked hybrid attention querying mechanism, optimizing multi-scale feature fusion, object localization, and instance-specific segmentation. First, this study presents a novel transformer-based masked hybrid attention mechanism using Laplacian query fusion on learnable query features and incorporating a novel tripartite sequence matching technique as part of the enhanced decoder block to improve the consistency of mask predictions and optimize decoder queries. The designed hybrid multi-head self- and cross-attention mechanisms aim to selectively integrate multi-scale features, ensuring optimal feature combinations for precise segmentation. Secondly, multiple class tokens are incorporated to improve object localization and capture class-specific characteristics within the transformer framework, leveraging the transformer decoders’ ability to learn distinct instance representations. Experimental results and extensive ablation studies demonstrate the effectiveness of the proposed approach on three publicly available datasets, obtaining better segmentation results compared to various state-of-the-art approaches using various evaluation metrics. Specifically, the proposed model achieves a Dice Score of 95.25%, 92.75%, and 85.25% on LUNA, ISIC, and DRIVE datasets, respectively. © The Author(s), under exclusive licence

关键词： Laplace transforms

来源：评论

学校读者我要写书评

暂无评论

Multiscale modelling and control of inductively coupled plasma etching based on recurrent neural networks

引用

International Journal of Wireless and Mobile Computing 2025年第1期28卷 34-46页

作者： Zhu, Lihong Chen, Qingchuan Nie, Junwei Liu, Kewei Department of Automation Engineering The Engineering and Technical College of Chengdu University of Technology Chengdu Sichuan Leshan China Southwestern Institute of Physics Sichuan Chengdu China Engineering Training Centre The Engineering and Technical College of Chengdu University of Technology Chengdu Sichuan Leshan China

In order to deepen the understanding of the process flow and improve the process, the plasma etching simulation is an essential means, but the calculation cost of the traditional simulation system is too high. Therefore, this study proposes a multi-scale model optimisation algorithm and model prediction and control system based on recurrent neural network. The multiscale model optimisation algorithm adopts a recursive neural network model to achieve simulation acceleration of the multi-scale model. The model prediction science system uses model reduction and recurrent neural network models to achieve full state prediction of macroscopic plasma models. The experimental results indicate that the simulation model can achieve the expected simulation results within 7.3 seconds. Moreover, the average etching depth of the model predictive control system shall not exceed 4.6%. Consequently, the proposed simulation model optimisation algorithm and model prediction and control system can effectively optimise and control the plasma etching. Copyright © 2025 Inderscience Enterprises Ltd.

关键词： Prediction models

来源：评论

学校读者我要写书评

暂无评论

Dual-head alternating attention-based adaptive convolutional framework for improving low-resource and imbalanced text classification

引用

Neural Computing and Applications 2025年 1-21页

作者： Agbesi, Victor Kwaku Chen, Wenyu Hossin, Md Altab Ukwuoma, Chiagoziem C. School of Computer Science and Engineering University of Electronic Science and Technology of China No. 2006 Xiyuan Ave West Hi-Tech Zone Sichuan Chengdu China School of Innovation and Entrepreneurship Chengdu University No. 2025 Chengluo Avenue 610106 Sichuan Chengdu China College of Nuclear Technology and Automation Engineering Chengdu University of Technology Sichuan Chengdu610059 China Sichuan Engineering Technology Research Center for Industrial Internet Intelligent Monitoring and Application Chengdu University of Technology Sichuan Chengdu610059 China

Low-resource imbalance texts pose significant challenges due to class imbalance, adversely affecting the performance of deep learning approaches. To address this issue, we propose a dual-head alternating attention-based adaptive convolutional framework with label embedding (AltXLM-AdaCLE), a novel framework designed for imbalanced text classification in low-resourced datasets. Our framework features a new alternating attention model, which combines a dual-head attention mechanism with a bidirectional long short-term memory network, facilitating dynamic filter generation that adapts to the input data’s characteristics and enhances the capture of relevant features and long-range dependencies. A key innovation in this study is the dual-head attention mechanism, which utilizes an alternating reading sequence to extract information from multiple document segments, thereby minimizing information loss and improving the understanding of complex relationships. Furthermore, we introduce a method for embedding class labels in a shared vector space with the text, allowing the model to leverage label information throughout the classification process, which is particularly beneficial for imbalanced datasets. Our framework also incorporates an adaptive convolutional layer that processes text features and label embeddings, generating more representative features of the underlying data distribution and improving classification performance. Experimental results on six low-resourced imbalance text datasets validate the effectiveness of AltXLM-AdaCLE with an average AUC gain of 26.42%, demonstrating its superiority over existing methods. © The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature 2025.

关键词： Vector spaces

来源：评论

学校读者我要写书评

暂无评论

Targets capture by distributed active swarms via bio-inspired reinforcement learning

引用

Science China(Physics,Mechanics & Astronomy) 2025年第1期68卷 206-217页

作者： Kun Xu Yue Li Jun Sun Shuyuan Du Xinpeng Di Yuguang Yang Bo Li Institute of Biomechanics and Medical Engineering Applied Mechanics LaboratoryDepartment of Engineering MechanicsTsinghua UniversityBeijing 100084China Institute of Nuclear and New Energy Technology Tsinghua UniversityBeijing 100084China Shanghai Aerospace Control Technology Institute Shanghai 201109China Shanghai Key Laboratory of Aerospace Intelligent Control Technology Shanghai 201109China College of Automation Engineering Nanjing University of Aeronautics and AstronauticsNanjing 210016China

Natural swarms arranged from cells to herds are usually decentralized but display intriguing collective intelligence in coordinating individuals across large scales to efficiently achieve their common *** from nature may provide new strategies for controlling collective dynamics of synthetic swarms to accomplish specific ***,we present a bioinspired computational framework that steers distributed active swarms to collectively capture and merge targets via reinforcement *** exploit collective milling structures of natural herds to cage the targets,and adopt a switching control policy inspired by sperms’chiral dynamics to optimize the trajectories of individuals,through which the active swarms can selforganize to enclose single or multiple distant targets in a dynamical,adaptive and scalable *** exists a critical swarm size,beyond which the excessive competition between agents would generate large mechanical forces,leading to capture instability but enabling the transition from short-distance to long-distance merging capture of multiple *** work provides physical insights into distributed active swarms and could offer a multilevel,decentralized strategy toward controlling swarm robotics in wide applications such as bio-medical devices,machine immunity,and target clearance.

关键词： swarm intelligence robotics multi-tasks artificial intelligence bionics

来源：评论

学校读者我要写书评

暂无评论

Security control for air-sea heterogeneous multiagent systems with cooperative-antagonistic interactions:An intermittent privacy preservation mechanism

引用

Science China(Technological Sciences) 2025年第4期68卷 179-192页

作者： Shoufeng YANG Hongjing LIANG Yingnan PAN Tieshan LI College of Control Science and Engineering Bohai UniversityJinzhou121013China School of Automation Engineering University of Electronic Science and Technology of ChinaChengdu611731China Laboratory of Electromagnetic Space Cognition and Intelligent Control Beijing100089China

antagonistic interactions,data security and transient-steady state performance of the system are two key *** ensure data security,an intermittent privacy preservation(IPP)mechanism is proposed for the first time.A novel setting time initial mask function and a novel intermittent mask function are *** can implement intermittent preservation for the system according to actual requirements,which solves the irreversibility problem after conventional mask disappears and balances control accuracy and system *** ensure transient-steady state performance,a novel error transformation function(ETF)is proposed and integrated into the predefined-time prescribed performance control *** to conventional hyperbolic tangent type ETFs,the proposed ETF can improve the convergence accuracy of errors under the same ***,a unified model of the air-sea HMASs is established,which improves the model accuracy compared with the simplified ***,the proposed IPP security control strategy is applied to the air-sea delivery mission to verify its feasibility and effectiveness.

关键词： air-sea heterogeneous multiagent systems(HMASs) security control intermittent privacy preservation(IPP) error transformation function(ETF)

来源：评论

学校读者我要写书评

暂无评论

STSR: An Efficient Model for Super-Resolution Reconstruction

STSR: An Efficient Model for Super-Resolution Reconstruction

引用

Consumer Electronics and Computer engineering (ICCECE), 2021 IEEE International Conference on

作者： Xinhui Liu College of Nuclear Technology and Automation Engineering Chengdu University of Technology Chengdu China

ISBN: (数字)9798331533113

ISBN: (纸本)9798331533120

STSR is an efficient and lightweight super-resolution model built upon the Swin Transformer architecture. By utilizing the hierarchical window-based self-attention mechanism of the Swin Transformer, STSR effectively captures global contextual information while ensuring high computational efficiency. To further enhance its performance and reduce model complexity, we incorporate lightweight design elements and residual structures. These improvements not only facilitate better gradient flow but also significantly boost the recovery of high-frequency details. Experimental results on the DIV2K dataset show that STSR delivers exceptional performance, significantly reducing both parameter count and computational costs.

关键词： Computational modeling Superresolution Computer architecture Transformers Computational efficiency Complexity theory Image reconstruction Consumer electronics

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：