检索结果-内蒙古大学图书馆

Chinese Control Conference (CCC)

作者： Lei Zhang Hao Xia Bao Cha School of Control Science and Control Engineering Dalian University of Technology Dalian P. R. China Science and Technology on Reactor System Design Technology Laboratory Nuclear Power Institute of China Chengdu P. R. China

The existing fault diagnosis methods based on deep neural networks mostly rely on process data for fault classification and diagnosis, which makes it difficult to distinguish minor faults with similar data features. Therefore, this paper proposes a deep learning-based fault diagnosis model that combines statistical indicators and neural networks. By analyzing and calculating the relevant information entropy to screen the statistical signal features of time domain and frequency domain, we can extract more hidden information from the raw data, and use the convolutional autoencoder to compress the statistical indicators for feature extraction, and use CNN network to extract features from raw data. Finally, the two sets of features are fused and input into the LSTM network to obtain temporal information for fault diagnosis. Tests conducted on the Tennessee Eastman (TE) process show that the proposed model outperforms typical classifiers and improves the diagnostic accuracy of minor faults.

关键词：

来源：评论

学校读者我要写书评

暂无评论

UDC: A Unified Neural Divide-and-Conquer Framework for Large-Scale Combinatorial Optimization Problems

arXiv

引用

arXiv 2024年

作者： Zheng, Zhi Zhou, Changliang Tong, Xialiang Yuan, Mingxuan Wang, Zhenkun School of System Design and Intelligent Manufacturing Southern University of Science and Technology Shenzhen China Huawei Noah’s Ark Lab Shenzhen China

Single-stage neural combinatorial optimization solvers have achieved near-optimal results on various small-scale combinatorial optimization (CO) problems without requiring expert knowledge. However, these solvers exhibit significant performance degradation when applied to large-scale CO problems. Recently, two-stage neural methods motivated by divide-and-conquer strategies have shown efficiency in addressing large-scale CO problems. Nevertheless, the performance of these methods highly relies on problem-specific heuristics in either the dividing or the conquering procedure, which limits their applicability to general CO problems. Moreover, these methods employ separate training schemes and ignore the interdependencies between the dividing and conquering strategies, often leading to sub-optimal solutions. To tackle these drawbacks, this article develops a unified neural divide-and-conquer framework (i.e., UDC) for solving general large-scale CO problems. UDC offers a Divide-Conquer-Reunion (DCR) training method to eliminate the negative impact of a sub-optimal dividing policy. Employing a high-efficiency Graph Neural Network (GNN) for global instance dividing and a fixed-length sub-path solver for conquering divided sub-problems, the proposed UDC framework demonstrates extensive applicability, achieving superior performance in 10 representative large-scale CO problems. The code is available at https://***/CIAM-Group/NCO_code/tree/main/single_objective/UDC-Large-scale-CO-master Copyright © 2024, The Authors. All rights reserved.

关键词： Combinatorial optimization

来源：评论

学校读者我要写书评

暂无评论

Evolution of heuristics: towards efficient automatic algorithm design using large language model 24

Evolution of heuristics: towards efficient automatic algorit...

引用

Proceedings of the 41st International Conference on Machine Learning

作者： Fei Liu Xialiang Tong Mingxuan Yuan Xi Lin Fu Luo Zhenkun Wang Zhichao Lu Qingfu Zhang Department of Computer Science City University of Hong Kong Huawei Noah's Ark Lab School of System Design and Intelligent Manufacturing Southern University of Science and Technology

Heuristics are widely used for dealing with complex search and optimization problems. However, manual design of heuristics can be often very labour extensive and requires rich working experience and knowledge. This paper proposes Evolution of Heuristic (EoH), a novel evolutionary paradigm that leverages both Large Language Models (LLMs) and Evolutionary Computation (EC) methods for Automatic Heuristic design (AHD). EoH represents the ideas of heuristics in natural language, termed thoughts. They are then translated into executable codes by LLMs. The evolution of both thoughts and codes in an evolutionary search framework makes it very effective and efficient for generating high-performance heuristics. Experiments on three widely studied combinatorial optimization benchmark problems demonstrate that EoH outperforms commonly used handcrafted heuristics and other recent AHD methods including FunSearch. Particularly, the heuristic produced by EoH with a low computational budget (in terms of the number of queries to LLMs) significantly outperforms widely-used human hand-crafted baseline algorithms for the online bin packing problem.

关键词：

来源：评论

学校读者我要写书评

暂无评论

DPN: decoupling partition and navigation for neural solvers of min-max vehicle routing problems 24

DPN: decoupling partition and navigation for neural solvers ...

引用

Proceedings of the 41st International Conference on Machine Learning

作者： Zhi Zheng Shunyu Yao Zhenkun Wang Xialiang Tong Mingxuan Yuan Ke Tang Department of Computer Science and Engineering Southern University of Science and Technology Shenzhen China Department of Computer Science City University of Hong Kong Hong Kong SAR and School of System Design and Intelligent Manufacturing Southern University of Science and Technology Shenzhen China School of System Design and Intelligent Manufacturing Southern University of Science and Technology Shenzhen China China Huawei Noah's Ark Lab Shenzhen China

The min-max vehicle routing problem (min-max VRP) traverses all given customers by assigning several routes and aims to minimize the length of the longest route. Recently, reinforcement learning (RL)-based sequential planning methods have exhibited advantages in solving efficiency and optimality. However, these methods fail to exploit the problem-specific properties in learning representations, resulting in less effective features for decoding optimal routes. This paper considers the sequential planning process of min-max VRPs as two coupled optimization tasks: customer partition for different routes and customer navigation in each route (i.e., partition and navigation). To effectively process min-max VRP instances, we present a novel attention-based Partition-and-Navigation encoder (P&N Encoder) that learns distinct embeddings for partition and navigation. Furthermore, we utilize an inherent symmetry in decoding routes and develop an effective agent-permutation-symmetric (APS) loss function. Experimental results demonstrate that the proposed Decoupling-Partition-Navigation (DPN) method significantly surpasses existing learning-based methods in both single-depot and multi-depot minmax VRPs. Our code is available at.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Direct Numerical Simulation of Turbulent Flow and Heat Transfer in a Rectangular Duct with High Aspect Ratio

SSRN

引用

SSRN 2023年

作者： Sun, Yumeng Li, Jie Wei, Zonglan Ai, Zhelun Miao, Heng Deng, Jian Sino-French Institute of Nuclear Engineering and Technology Sun Yat-sen University Guangdong Zhuhai519082 China Science and Technology on Reactor System Design Technology Laboratory Nuclear Power Institute of China Sichuan Chengdu610213 China

Plate-type fuel is widely used due to its strong heat transfer capability, and its narrow rectangular geometry has different effects on flow heat transfer than traditional rectangular channels. Therefore, in this work, In this work, Direct Numerical Simulation(DNS) of a narrow rectangular duct was performed with the aim of analyzing flow and heat transfer phenomena and constructing a database of turbulent flow. The open-source program Nek5000 based on the spectral element method is used, the geometry is 0.57×0.00367×0.05 m with a high Aspect Ratio(AR) of 13.6, and the bulk Reynolds number is 24000 with a heat flux of 560 kW/m2. Common turbulence statistics like the time mean field, reynolds stress, turbulent heat flux distribution profile, and turbulent kinetic energy budget are obtained. The distribution of Reynolds stress and turbulent kinetic energy budget was analyzed, and the impact of high aspect ratio and material properties on flow and heat transfer was investigated. Additionally, the effects of single-sided and double-sided heating were studied. The turbulent flow database obtained from this study will be utilized for the later evaluation of turbulence models. © 2023, The Authors. All rights reserved.

关键词： Direct numerical simulation

来源：评论

学校读者我要写书评

暂无评论

In-situ Thermal Oxidation of Fusion PFM Tungsten Using Atmospheric Environmental TEM

引用

Microscopy and microanalysis : the official journal of Microscopy Society of America, Microbeam Analysis Society, Microscopical Society of Canada 2023年第1期29卷 1462-1463页

作者： Sainju, Rajat Togaru, Maanas Zhang, Lichun Jiang, Weilin Setyawan, Wahyu Atwani, Osman El Zhu, Yuanyuan Department of Materials Science and Engineering University of Connecticut Storrs CT United States Institute of Materials Science University of Connecticut Storrs CT United States Reactor Materials and Mechanical Design Group Nuclear Sciences Division Pacific Northwest National Laboratory Richland WA United States Materials Science and Technology Los Alamos National Lab Los Alamos NM United States

来源：评论

学校读者我要写书评

暂无评论

Theoretical research on two-phase flow instability in parallel rectangular channels under periodic perturbation

Theoretical research on two-phase flow instability in parall...

引用

2020 International Conference on Nuclear Engineering, ICONE 2020, collocated with the ASME 2020 Power Conference

作者： Qian, Libo Deng, Jian Huang, Tao Cai, Rong Science and Technology on Reactor System Design Technology Laboratory Npic Chengdu Sichuan China

ISBN: (纸本)9780791883785

A theoretical model for Density Wave Oscillations (DWOs) flow instability in parallel rectangular channels under periodic heaving motion is established with a lumped mathematical model based on homogenous hypothesis. The parallel rectangular channels comprise of the entrance section, the heating section, the riser section and the upper- and lower plenums, which guarantee the isobaric pressure drop condition between channels and the model consists of boiling channel model, pressure drop model, parallel channel model, additional pressure drop model generated by heaving motions, the constitutive and numerical models. The effect of periodic perturbation is introduced through additional pressure drop in the momentum equation. The model is validated with experimental data of a twin-rectangular-channel flow instability experiment under static condition. Then the flow instability in parallel-rectangular-channel system is studied under periodic perturbation and the margin of flow instability and the threshold power of the system under static condition is calculated as basis condition for comparison. The effect of the amplitude and period of perturbation is analyzed analytically and the results show that the amplitude and period of perturbation shows little effect on flow instability. While when the additional pressure difference introduced by heaving motion is comparable with that under static condition, the effect of amplitude becomes stronger. And the period of perturbation strongly effects the threshold power when it is identical to that of natural period of the system, which can be explained by resonance between the perturbation and the system. And this effect is even stronger when the asymmetric heating condition is introduced. © 2020 American Society of Mechanical Engineers (ASME). All rights reserved.

关键词： Pressure drop

来源：评论

学校读者我要写书评

暂无评论

Experimental study on the sub-channel void fraction characteristics of bubbly flow in rod bundles

Experimental study on the sub-channel void fraction characte...

引用

2020 International Conference on Nuclear Engineering, ICONE 2020, collocated with the ASME 2020 Power Conference

作者： Ren, Quan-Yao Pu, Zeng-Ping Zheng, Mei-Yin Su, Min Chen, Ping Pan, Liang-Ming He, Hui He, Qing-Che Science and Technology on Reactor System Design Technology Laboratory Chengdu China Chongqing China

ISBN: (纸本)9780791883785

The gas-liquid two-phase flow behaviors are always associated with its dynamic void fraction, such as flow resistance, heat transfer coefficient, phase distribution, critical heat flux etc. As regard to the commercial PWR and BWR, rod bundles are the typical geometry, which contains many subchannels for coolant flowing. In present study, the sub-channel void fraction was measured in 5×5 rod bundles with the subchannel impedance void meter consisting of 12 strip electrodes. Based on the measured void fraction in different sub-channels, the void fraction dynamics, PDF (probability distribution function) and CDF (cumulative distribution function) curves were analyzed to make clear the effect of superficial gas and liquid velocity, flow development and casing tube. The empirical correlation for PDF of dynamic sub-channel void fraction has been developed, which showed good fitness with PDF and CDF curves and satisfying accuracy of averaged void fraction. © 2020 American Society of Mechanical Engineers (ASME). All rights reserved.

关键词： Void fraction

来源：评论

学校读者我要写书评

暂无评论

An Improved Method for Underwater Image Super-Resolution and Enhancement 4

An Improved Method for Underwater Image Super-Resolution and...

引用

4th IEEE International Conference on Electronics technology, ICET 2021

作者： Li, Hao Zhang, Changhua Wan, Naijia Chen, Qian Wang, Donghun Song, Danrong University of Electronic Science and Technology of China School of Mechanical and Electrical Engineering Chengdu China University of Electronic Science and Technology of China School of Information and Software Engineering Chengdu China Technology Laboratory Nuclear Power Institute of China Science and Technology on Reactor System Design Chengdu China

ISBN: (纸本)9781728176734

Underwater object detection and recognition are still challenging because of the degradation caused by the complex underwater imaging environment and lighting conditions. Aiming at the problems of the simultaneous enhancement and super-resolution (SESR), a method with single image super-resolution and enhancement of underwater imagery is proposed. Firstly, an underwater image super-resolution algorithm is proposed to remove blurring, and then an end-to-end underwater image enhancement network is used to compensate for color casts and produce natural color enhancement images. In addition, we verify the effectiveness of the proposed model through qualitative analysis and quantitative experiments and compare the performance with several state-of-the-art models, which show that the proposed method performs better. © 2021 IEEE.

关键词： Image enhancement

来源：评论

学校读者我要写书评

暂无评论

Understanding Xenon and Vacancy Behaviors in Uo2, Un and U3si2: A Comparative Ab Initio Study

SSRN

引用

SSRN 2022年

作者： Zhao, Jiajun Sun, Dan Wang, Yuanyuan Xi, Liu Chen, Ping Zhao, Jijun Dalian University of Technology Liaoning Dalian China Science and Technology on Reactor System Design Technology Laboratory Nuclear Power Institute of China Sichuan Chengdu China

Extensive attention has been paid to accident tolerant fuels (ATFs), such as uranium mononitride (UN) and uranium sesquisilicide (U3Si2), which are regarded as the potential candidates to replace uranium dioxide (UO2) in light water reactor (LWR). However, the thermodynamic behavior of fission gas atom in these fuels that can quantitatively affect the burnup characteristics of ATF needs to be explored. To this end, the systematic ab initio investigation on the energetic properties of xenon (Xe)-vacancy complexes in UO2, UN and U3Si2 is performed with the GGA+U approach as well as the corrected chemical potential. The stabilities of Xe-vacancy clusters, including interstitial trap site (IS), mono-, bi- and tri-atomic vacancy, are thoroughly assessed. The formation energies of vacancy complexes indicate that it is more likely to form vacancy cluster defects and their complexes with Xe in UO2 and to generate mono-atomic vacancy and Xe-vacancy complex in both UN and U3Si2. Xe can be strictly confined by the trap sites in UO2 and UN, and yet in U3Si2, it prefers to move to the centre of a large free volume trap site. The strongest soluble characteristic of Xe in U3Si2 indicates excellent storage compacity of fission gas products in the matrix. This work can provide comprehensive insights into the original reason of interplay between Xe and vacancy as well as the thermodynamic behavior of those defects in the uranium-based fuels. © 2022, The Authors. All rights reserved.

关键词： Uranium dioxide

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：