Search | Engineering

订阅投稿

首页工程期刊工程焦点工程成就工程前沿关于我们 English

资源类型

期刊论文 528

年份

2023 83

2022 89

2021 50

2020 53

2019 40

2018 26

2017 25

2016 18

2015 15

2014 10

2013 6

2012 2

2011 14

2010 21

2009 14

2008 15

2007 13

2006 3

2005 1

2003 2

展开︾

关键词

机器学习 27

深度学习 15

人工智能 10

代理模型 2

强化学习 2

材料设计 2

结构健康监测 2

高层建筑 2

2D—3D配准 1

3D打印 1

6G 1

ARM 1

CAN总线 1

CCUS 1

CO₂利用 1

COVID-19 1

Cuk矩阵变换器 1

GaAs基微结构材料 1

H∞控制；零和动态博弈；强化学习；自适应动态规划；极小极大Q-学习；策略迭代 1

展开︾

检索范围：

排序：展示方式：

MSWNet: A visual deep machine learning method adopting transfer learning based upon ResNet 50 for municipal

《环境科学与工程前沿（英文）》 2023年第17卷第6期 doi: 10.1007/s11783-023-1677-1

摘要：

● MSWNet was proposed to classify municipal solid waste.

关键词： Municipal solid waste sorting Deep residual network Transfer learning Cyclic learning rate Visualization

HTML PDF 收藏

Modeling of oil near-infrared spectroscopy based on similarity and transfer learning algorithm

Yifei Wang, Kai Wang, Zhao Zhou, Wenli Du

《化学科学与工程前沿（英文）》 2019年第13卷第3期页码 599-607 doi: 10.1007/s11705-019-1807-2

摘要： Near-infrared spectroscopy mainly reflects the frequency-doubled and total-frequency absorption information of hydrogen-containing groups (O‒H, C‒H, N‒H, S‒H) in organic molecules for near-infrared lights with different wavelengths, so it is applicable to testing of most raw materials and products in the field of petrochemicals. However, the modeling process needs to collect a large number of laboratory analysis data. There are many oil sources in China, and oil properties change frequently. Modeling of each raw material is not only unfeasible but also will affect its engineering application efficiency. In order to achieve rapid modeling of near-infrared spectroscopy and based on historical data of different crude oils under different detection conditions, this paper discusses about the feasibility of the application of transfer learning algorithm and makes it possible that transfer learning can assist in rapid modeling using certain historical data under similar distributions under a small quantity of new data. In consideration of the requirement of transfer learning for certain similarity of different datasets, a transfer learning method based on local similarity feature selection is proposed. The simulation verification of spectral data of 13 crude oils measured by three different probe detection methods is performed. The effectiveness and application scope of the transfer modeling method under different similarity conditions are analyzed.

关键词： near-infrared spectroscopy transfer learning similarity modeling

HTML PDF 收藏

一种用于工业过程监测的鲁棒迁移字典学习算法 Article

阳春华, 梁慧平, 黄科科, 李勇刚, 桂卫华

《工程（英文）》 2021年第7卷第9期页码 1262-1273 doi: 10.1016/j.eng.2020.08.028

摘要：

由于数据驱动的过程监测方法具有普遍性，且不依赖反应机理，其已经成为复杂工业系统过程监测的主流。然而，大多数数据驱动的过程监测方法均假设历史训练数据和在线测试数据遵循相同的分布。事实上，由于工业系统恶劣的环境，从实际工业过程中收集的数据总是受到许多因素的影响，如多变的操作环境、原材料的变化和生产指标的修改。这些因素通常会使在线监测数据和历史训练数据分布不同，从而导致过程监测任务中的模型失配。因此，当将从训练数据中学习的模型应用于实际的在线监测时，很难实现精确的过程监测。为了解决操作环境变化导致的历史训练数据和在线测试数据之间的分布差异问题，提出了一种鲁棒的迁移字典学习（RTDL）算法用于工业过程监测。RTDL是表示学习和域自适应迁移学习的协同方法。该方法将历史训练数据和在线测试数据分别作为迁移学习问题的源域和目标域。然后将最大均值差异正则化和线性判别分析正则化引入字典学习框架，可以减少源域和目标域之间的分布差异。这样，即使源域和目标域的特征在实际变化的操作环境的干扰下明显不同，仍可以学习鲁棒的字典。这样的字典可以有效地提高过程监测和模态识别的性能。通过数值仿真和两个工业系统的实验验证了该方法的有效性和优越性。

关键词：过程监测多模态过程字典学习迁移学习

HTML PDF 收藏

Dynamic simulation of gas turbines via feature similarity-based transfer learning

Dengji ZHOU, Jiarui HAO, Dawen HUANG, Xingyun JIA, Huisheng ZHANG

《能源前沿（英文）》 2020年第14卷第4期页码 817-835 doi: 10.1007/s11708-020-0709-9

摘要： Since gas turbine plays a key role in electricity power generating, the requirements on the safety and reliability of this classical thermal system are becoming gradually strict. With a large amount of renewable energy being integrated into the power grid, the request of deep peak load regulation for satisfying the varying demand of users and maintaining the stability of the whole power grid leads to more unstable working conditions of gas turbines. The startup, shutdown, and load fluctuation are dominating the operating condition of gas turbines. Hence simulating and analyzing the dynamic behavior of the engines under such instable working conditions are important in improving their design, operation, and maintenance. However, conventional dynamic simulation methods based on the physic differential equations is unable to tackle the uncertainty and noise when faced with variant real-world operations. Although data-driven simulating methods, to some extent, can mitigate the problem, it is impossible to perform simulations with insufficient data. To tackle the issue, a novel transfer learning framework is proposed to transfer the knowledge from the physics equation domain to the real-world application domain to compensate for the lack of data. A strong dynamic operating data set with steep slope signals is created based on physics equations and then a feature similarity-based learning model with an encoder and a decoder is built and trained to achieve feature adaptive knowledge transferring. The simulation accuracy is significantly increased by 24.6% and the predicting error reduced by 63.6% compared with the baseline model. Moreover, compared with the other classical transfer learning modes, the method proposed has the best simulating performance on field testing data set. Furthermore, the effect study on the hyper parameters indicates that the method proposed is able to adaptively balance the weight of learning knowledge from the physical theory domain or from the real-world operation domain.

关键词： gas turbine dynamic simulation data-driven transfer learning feature similarity

HTML PDF 收藏

种基于特征选择与迁移学习的度量补偿软件缺陷预测方法 Research Article

陈锦富1,2,王小丽1,2,蔡赛华1,2,徐家平1,陈静怡1,陈海波1

《信息与电子工程前沿（英文）》 2022年第23卷第5期页码 715-731 doi: 10.1631/FITEE.2100468

摘要：跨项目软件缺陷预测解决了传统缺陷预测中训练数据不足的问题，克服了将多个不同源项目中学习的模型应用于目标项目的挑战。与此同时，出现两个新问题：（1）模型训练过程中过多无关和冗余特征影响训练效率，降低了模型预测精度；（2）由于开发环境等因素，度量值的分布因项目而异，当模型用于跨项目预测时，预测精度较低。本文引入皮尔逊特征选择方法解决数据冗余问题，采用基于迁移学习的度量补偿技术解决源项目和目标项目之间数据分布差异较大的问题。提出一种基于特征选择和迁移学习的度量补偿软件缺陷预测方法。实验结果表明，用该方法构建的模型在AUC（接收器工作特性曲线下面积）值和F1度量指标上取得较好结果。

关键词：缺陷预测；特征选择；迁移学习；度量补偿

HTML PDF 收藏

基于两级层次特征学习的图像分类方法 Article

Guang-hui SONG,Xiao-gang JIN,Gen-lang CHEN,Yan NIE

《信息与电子工程前沿（英文）》 2016年第17卷第9期页码 897-906 doi: 10.1631/FITEE.1500346

摘要：概要：在图像分类任务中，不同类别之间的相似度是不同的，样本经常被误分到相似度较高的类别中。为了区分高度相似类别中的样本，需要更加具体的图像特征，以便于分类器能够提高分类性能。本文提出了一种新颖、有效的基于深度卷积神经网络的两级层次特征学习框架。首先，不同层次的深度特征抽取器使用迁移学习方法进行训练。然后，从全部类别中抽取的通用特征和从高度相似类别中抽取的具体特征被融合成一个特征向量，并将其输入线性分类器进行分类。最后，基于Caltech-256、Oxford Flower-102和Tasmania Coral Point Count三个图像数据集的实验证明，通过两级层次特征学习的深度特征的表达能力十分强大，与传统的扁平多分类方法相比，我们提出的方法能有效的提高分类精度。

关键词：迁移学习；特征学习；深度卷积神经网络；层次分类；谱聚类

HTML PDF 收藏

Deep convolutional neural network for multi-level non-invasive tunnel lining assessment

《结构与土木工程前沿（英文）》 2022年第16卷第2期页码 214-223 doi: 10.1007/s11709-021-0800-2

摘要： In recent years, great attention has focused on the development of automated procedures for infrastructures control. Many efforts have aimed at greater speed and reliability compared to traditional methods of assessing structural conditions. The paper proposes a multi-level strategy, designed and implemented on the basis of periodic structural monitoring oriented to a cost- and time-efficient tunnel control plan. Such strategy leverages the high capacity of convolutional neural networks to identify and classify potential critical situations. In a supervised learning framework, Ground Penetrating Radar (GPR) profiles and the revealed structural phenomena have been used as input and output to train and test such networks. Image-based analysis and integrative investigations involving video-endoscopy, core drilling, jacking and pull-out testing have been exploited to define the structural conditions linked to GPR profiles and to create the database. The degree of detail and accuracy achieved in identifying a structural condition is high. As a result, this strategy appears of value to infrastructure managers who need to reduce the amount and invasiveness of testing, and thus also to reduce the time and costs associated with inspections made by highly specialized technicians.

关键词： concrete structure GPR damage classification convolutional neural network transfer learning

HTML PDF 收藏

在非对称大规模MIMO系统中基于集成—迁移学习的信道参数预测 Research Article

何遵文1,李悦1,张焱1,张万成1,张恺恩1,郭柳1,王海明2

《信息与电子工程前沿（英文）》 2023年第24卷第2期页码 275-288 doi: 10.1631/FITEE.2200169

摘要：近年来，多智能体深度强化学习（multi-agent deep 为降低第六代移动网络中的数据处理负担和硬件成本，非对称大规模多入多出（multiple-input multiple-output，MIMO）系统被提出。然而，在非对称大规模MIMO系统中，上行和下行无线信道之间的互易性是无效的。因此，需要基站和用户设备都发送导频来预测双向信道，这会消耗更多传输和计算资源。本文提出一种基于集成迁移学习的非对称大规模MIMO系统的信道参数预测方法，可以预测多个下行信道参数，包括路径损耗、多径数、时延扩展和角度扩展。选择上行信道参数和环境特征来预测下行参数。此外，提出一种基于SHAP（SHapley Additive exPlanations）值和最小描述长度标准的两步特征选择算法，以降低由弱相关或不相关特征引起的计算复杂度和对模型准确性的负面影响。引入实例迁移方法，以支持预测模型应对在新的传播条件下难以在短时间内收集足够训练数据的问题。仿真结果表明，该方法比反向传播神经网络和3GPP TR 38.901信道模型更准确。当波束宽度或通信扇区发生变化时，所提出的基于实例迁移的方法在预测下行参数方面优于没有迁移学习的方法。

关键词：非对称大规模MIMO系统；信道模型；集成学习；实例迁移；参数预测

HTML PDF 收藏

一个格上不经意传输协议的量子安全性分析 Article

Mo-meng LIU, Juliane KRÄMER, Yu-pu HU, Johannes BUCHMANN

《信息与电子工程前沿（英文）》 2017年第18卷第9期页码 1348-1369 doi: 10.1631/FITEE.1700039

摘要：不经意传输协议（oblivious transfer, OT）因其简易的密码功能广泛应用于安全多方计算。

关键词：不经意传输；后量子；格公钥；带差错学习；通用可复合

HTML PDF 收藏

基于迁移学习与多视角感兴趣点的膝关节运动追踪网络 Article

王聪, 谢帅宁, 李康, 王重阳, 刘旭东, 赵亮, 蔡宗远

《工程（英文）》 2021年第7卷第6期页码 881-888 doi: 10.1016/j.eng.2020.03.016

摘要：

近年来，深度学习为一种基于二维（2D）—三维（3D）配准技术以测量人体膝关节运动的方法，该方法提供了快速完成配准并增加捕捉范围的可能性。但这类方法受限于大量的数据需求，因此，我们提出了一种基于特征的迁移学习法，用于提取荧光透视影像的特征。通过三个受试者以及不到100对荧光透视影像，我们获得了40%的平均配准成功率。本研究提出的基于学习的配准方法，可在荧光透视影像数量有限时使用。

关键词： 2D—3D配准机器学习领域自适应点对应

HTML PDF 收藏

Augmentation of natural convective heat transfer by acoustic cavitation

Jun CAI, Xiulan HUAI, Shiqiang LIANG, Xunfeng LI,

《能源前沿（英文）》 2010年第4卷第3期页码 313-318 doi: 10.1007/s11708-009-0064-3

摘要： An experimental study was conducted to investigate the effects of acoustic cavitation on natural convective heat transfer from a horizontal circular tube. The experimental results indicated that heat transfer could be enhanced by acoustic cavitation and had the best effect when the head of the ultrasonic transducer was over the midpoint of the circular tube, and the distance between the head and the tube equaled 15 mm. The augmentation at low heat flux was better than that in the case of high heat flux. Based on experimental results, the correlation formula of Nusselt number for water was obtained.

关键词： heat transfer enhancement augmentation acoustic cavitation acoustic streaming convective heat transfer

HTML PDF 收藏

Experimental study of heat transfer coefficient with rectangular baffle fin of solar air heater

Foued CHABANE,Nesrine HATRAF,Noureddine MOUMMI

《能源前沿（英文）》 2014年第8卷第2期页码 160-172 doi: 10.1007/s11708-014-0321-y

摘要： This paper presents an experimental analysis of a single pass solar air collector with, and without using baffle fin. The heat transfer coefficient between the absorber plate and air can be considerably increased by using artificial roughness on the bottom plate and under the absorber plate of a solar air heater duct. An experimental study has been conducted to investigate the effect of roughness and operating parameters on heat transfer. The investigation has covered the range of Reynolds number from 1259 to 2517 depending on types of the configuration of the solar collectors. Based on the experimental data, values of Nusselt number have been determined for different values of configurations and operating parameters. To determine the enhancement in heat transfer and increment in thermal efficiency, the values of Nusselt have been compared with those of smooth duct under similar flow conditions.

关键词： Nusselt number flow rate heat transfer heat transfer coefficient thermal efficiency forced convection

HTML PDF 收藏

Effect of heat transfer space non-uniformity of combustion chamber components on in-cylinder heat transfer

Jizu LV, Minli BAI, Long ZHOU, Jian ZHOU,

《能源前沿（英文）》 2010年第4卷第3期页码 392-401 doi: 10.1007/s11708-009-0066-1

摘要： Combustion chamber components (cylinder head-cylinder liner-piston assembly-oil film) were treated as a coupled body. Based on the three-dimensional numerical simulation of the heat transfer of the coupled body, a coupled three-dimensional calculation model for the in-cylinder working process and the combustion chamber components was built with domain decomposition and boundary coupling method, which adopts the coupled three-dimensional simulation of in-cylinder working process and the combustion chamber components. The model was applied in the investigation of the influence of space non-uniformity in heat transfer among combustion chamber components on in-cylinder heat transfer. The results show that the effect of wall temperature space non-uniform distribution of combustion chamber components on heat transfer happens mainly at the end of the compression stroke and expansion stroke. Therefore, it can be concluded that wall temperature space non-uniform distribution of combustion chamber components would influence heat transfer during the intake and exhaust stroke obviously.

关键词： heat transfer space non-uniformity soot emission in-cylinder diesel

HTML PDF 收藏

Heat transfer coefficient of wheel rim of large capacity steam turbines

SHI Jinyuan, DENG Zhicheng, YANG Yu, JUN Ganwen

《能源前沿（英文）》 2008年第2卷第1期页码 20-24 doi: 10.1007/s11708-008-0015-4

摘要： A way of calculating the overall equivalent heat transfer coefficient of wheel rims of large capacity steam turbines is presented. The method and formula to calculate the mean forced convection heat-transfer coefficient of the surface of the blade and for the bottom wall of the blade passage, are introduced. The heat transmission from the blade to the rim was simplified by analogy to heat transmission in the fins. A fin heat transfer model was then used to calculate the equivalent heat transfer coefficient of the blade passage. The overall equivalent heat transfer coefficient of the wheel rim was then calculated using a cylindrical surface model. A practical calculation example was presented. The proposed method helps determine the heat transfer boundary conditions in finite element analyses of temperature and thermal stress fields of steam turbine rotors.

关键词： convection heat-transfer capacity heat-transfer coefficient bottom transmission

HTML PDF 收藏

Review of the LNG intermediate fluid vaporizer and its heat transfer characteristics

《能源前沿（英文）》 2022年第16卷第3期页码 429-444 doi: 10.1007/s11708-021-0747-y

摘要： The intermediate fluid vaporizer (IFV), different from other liquefied natural gas (LNG) vaporizers, has many advantages and has shown a great potential for future applications. In this present paper, studies of IFV and its heat transfer characteristics in the LNG vaporization unit E2 are systematically reviewed. The research methods involved include theoretical analysis, experimental investigation, numerical simulation, and process simulation. First, relevant studies on the overall calculation and system design of IFV are summarized, including the structural innovation design, the thermal calculation model, and the selection of different intermediate fluids. Moreover, studies on the fluid flow and heat transfer behaviors of the supercritical LNG inside the tubes and the condensation heat transfer of the intermediate fluid outside the tubes are summarized. In the thermal calculations of the IFV, the selections of the existing heat transfer correlations about the intermediate fluids are inconsistent in different studies, and there lacks the accuracy evaluation of those correlations or comparison with experimental data. Furthermore, corresponding experiments or numerical simulations on the cryogenic condensation heat transfer outside the tubes in the IFV need to be further improved, compared to those in the refrigeration and air-conditioning temperature range. Therefore, suggestions for further studies of IFV are provided as well.

关键词： intermediate fluid vaporizer design of structure and intermediate fluid condensation heat transfer