检索结果-南通市图书馆

在线全文

学校读者我要写书评

暂无评论

基于增强迁移学习的轴承故障诊断研究

中国工程机械学报 2024年第4期22卷 534-539页

作者：包从望车守全刘永志邹静六盘水师范学院矿业与机械工程学院贵州六盘水553000

深度迁移学习方法常被用于轴承故障迁移诊断,由于未充分考虑目标域对决策边界的影响,当2个领域分布差异较大时易导致迁移任务失败。针对此不足,提出了一种增强迁移学习方法(ETLM)用于轴承故障诊断,以改进轴承在变工况下的故障诊断性能... 详细信息

深度迁移学习方法常被用于轴承故障迁移诊断,由于未充分考虑目标域对决策边界的影响,当2个领域分布差异较大时易导致迁移任务失败。针对此不足,提出了一种增强迁移学习方法(ETLM)用于轴承故障诊断,以改进轴承在变工况下的故障诊断性能。该方法通过构建双层网络模型结构,以源域训练模型1为基础,并将权重参数结果复制到模型2中,在模型2中通过引入目标域样本的伪标签自适应调整决策边界。为提高样本的识别效果,采用2个独立分类器的结构训练模型1和模型2,训练过程中以最大化分类差异、最小化源域和目标域分布差异为目标迭代更新权重参数。将交叉熵损失函数、最大均值差异(MMD)、分类判别损失函数作为优化准则,以减小2个领域间的分布差异,增强迁移学习能力。通过轴承故障诊断实验台数据对所提方法进行分析,结果表明,ETLM方法在特征学习、识别率和迁移效果方面明显优于其他几种方法。

关键词：故障诊断迁移学习增强迁移学习滚动轴承

Endocrine therapy of short duration prevents local and contralateral recurrence of ductal carcinoma in situ of the breast: A multicenter retrospective cohort study in China

在线全文

学校读者我要写书评

暂无评论

中华医学杂志(英文版) 2024年第14期137卷 1756-1758页

作者： Wang Zhen Jin Zining Zhang Huanrui Xu Guiying Zhang Dianlong Fang Fengqi Xing Hua Wang Jia Guo Baoliang Zhu Guolian liu yongzhi Li Jieqing Cai Zhengang Sun Liang Zhang Yuting Zhou Tianyang liu Chang Zhou Baosen Jin Feng Zhang Yan Song Dong Chen Bo Department of Breast Surgery The First Hospital of China Medical University Shenyang Liaoning China Department of Breast and Thyroid Surgery Shandong Provincial Hospital Affiliated to Shandong First Medical University Jinan Shandong China Department of Geriatric Medicine The First Hospital of China Medical University Shenyang Liaoning China Department of Breast Surgery Jilin Cancer Hospital Changchun Jilin China Department of Breast and Thyroid Surgery Affiliated Zhongshan Hospital of Dalian University Dalian Liaoning China Department of Oncology The First Affiliated Hospital of Dalian Medical University Dalian Liaoning China Department of Breast Surgery China-Japan Union Hospital Affiliated to Jilin University Changchun Jilin China Department of Breast Surgery The Second Hospital of Dalian Medical University Dalian Liaoning China Department of Breast Surgery The Second Affiliated Hospital of Harbin Medical University Harbin Heilongjiang China Department of Breast Surgery The Fifth People’s Hospital of Shenyang Shenyang Liaoning China Department of Breast Surgery General Hospital of Benxi Iron &amp Steel Industry Group of Liaoning Health Industry Group Benxi Liaoning China Department of Breast Surgery The Fourth People’s Hospital of Shenyang Shenyang Liaoning China Department of Breast Surgery The First Affiliated Hospital of Dalian Medical University Dalian Liaoning China Department of Clinical Epidemiology and Center of Evidence Based Medicine The First Hospital of China Medical University Shenyang Liaoning China Department of Breast Surgery The Second Hospital of Jilin University Changchun Jilin China Department of Breast Surgery General Surgery Center The First Hospital of Jilin University Changchun Jilin China

To the Editor: De-escalating the treatment of ductal carcinoma in situ (DCIS) of the breast has been investigated in recent years. A 5-year adjuvant endocrine therapy (AET) regimen is the recommended treatment for hormone receptor-positive DCIS. Selective estrogen receptor modulators (such as the commonly used tamoxifen and toremifene) and aromatase inhibitors (AIs) are the currently used AET drugs. However, long-term AET is accompanied by severe side effects such as menopausal symptoms, endometrial carcinoma, cardiovascular events, venous thromboembolic events, osteoporosis, and joint pain that can significantly reduce patients’ quality of life. It was reported that only 50% of the patients completed the recommended 5-year AET mainly because of side effects, while higher adherence rates were observed with shorter-duration regimens. Therefore, de-escalation of AET should be considered to reduce the side effects of AET and improve adherence.

关键词： ADHERENCE

经验性与人工智能指导下精准肺段切除术效果比较的回顾性队列研究

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

中国胸心血管外科临床杂志 2024年第10期31卷 1422-1427页

作者：陈剑詹必成汤勇刘永志李根水刘建安庆市立医院心胸外科安徽安庆246003

目的比较通过薄层CT二维图像进行胸腔镜下经验性肺段切除手术和通过人工智能(artificial intelligence,AI)软件进行规划的精准肺段切除手术的临床效果,为临床开展肺段切除术提供一些参考。方法回顾性分析2019—2022年在安庆市立医院心... 详细信息

目的比较通过薄层CT二维图像进行胸腔镜下经验性肺段切除手术和通过人工智能(artificial intelligence,AI)软件进行规划的精准肺段切除手术的临床效果,为临床开展肺段切除术提供一些参考。方法回顾性分析2019—2022年在安庆市立医院心胸外科完成的胸腔镜下肺段手术患者的临床资料。将2019年1月—2021年9月经验性肺段切除手术患者纳入A组,2021年10月—2022年12月精准肺段切除手术患者纳入B组。比较两组术前Hookwire定位针用量、满足肿瘤学标准患者比例、手术时间、术中出血量、术后胸腔引流时间、术后住院时间和中转开胸率。结果共纳入患者322例。A组共158例,其中男56例、女102例,平均年龄(56.86±8.82)岁。B组共164例,其中男55例、女109例,平均年龄(56.69±9.05)岁。所有患者均顺利在胸腔镜下完成肺段切除手术,切缘不能满足肿瘤学标准患者进一步行扩大范围切除甚至肺叶切除手术,无围术期死亡。A组肺段切除使用定位针数量多于B组[47(29.7%)vs.9(5.5%),P0.05)。结论术前借助AI软件进行手术规划,能有效指导胸腔镜下解剖性肺段切除,在保证结节切缘符合肿瘤学要求的同时能显著减少肺结节定位针的数量。

关键词：人工智能肺段切除术肺结节术前定位

保理合同中将有应收账款的转让标准--基于232份生效判决书的实证分析

在线全文

学校读者我要写书评

暂无评论

镇江高专学报 2024年第2期37卷 52-57页

作者：刘永智陈慎谦江苏财会职业学院财税学院江苏连云港222061 淮安市中级人民法院民三庭江苏淮安223001

在司法实践中,将有应收账款的转让标准存在争论,判决结果不一。通过实证分析发现,法院判断将有应收账款是否符合对外转让条件存在主体、客体与形式3类标准。结合将有应收账款的特征,明确将有应收账款主体标准为任一民事主体,统一客体标... 详细信息

在司法实践中,将有应收账款的转让标准存在争论,判决结果不一。通过实证分析发现,法院判断将有应收账款是否符合对外转让条件存在主体、客体与形式3类标准。结合将有应收账款的特征,明确将有应收账款主体标准为任一民事主体,统一客体标准为合理期待性,完善形式标准即建立预先通知与转让登记制度。

关键词：应收账款财产价值预先通知转让登记

Numeric modelling and risk assessment of pollutions in the Chinese Bohai Sea

在线全文

学校读者我要写书评

暂无评论

Science China Earth Sciences 2017年第8期60卷 1546-1557页

作者： liu yongzhi SHEN YouLi LV XianQing liu Qiang Key Laboratory of Physical Oceanography Ocean University of China Beihai Marine Environmental Monitoring Centre Station of State Oceanic Administration College of Engineering Ocean University of China

To simulate the dynamic process of total nitrogen(TN) in seas, numerical modelling combined with the adjoint method is implemented in this study. Because nonpoint source terms(ST) and initial values(IV) of TN are essential but difficult to determine, the adjoint method was applied to a numerical model, and the ST and IV terms of TN were inverted via routine monitoring data in the Bohai Sea. In twin experiments, the adjoint method was capable of inverting the prescribed spatio-temporally distributed ST and the spatial distributed IV. In practical experiments, the results demonstrated that the simulation precision with ST inversion was higher than that with IV inversion and was accurate with joint initial values and source term(IST) inversion. This result indicates that nonpoint source TN is essential for the simulation of TN concentration. Furthermore, the simulated results indicate that the pollution in three bays of the Bohai Sea is rather severe. The model in this study is not specific to the Bohai Sea and can be generalized to other areas, such as the Beibu Gulf. These findings may assist in the development of cost-effective controls for accidental or planned industrial pollutant releases into coastal waters.

关键词： Risk assessment Adjoint method Initial value Numeric modelling Bohai Sea of China

Water characteristics of abyssal and hadal zones in the southern Yap Trench observed with the submersible Jiaolong

在线全文

学校读者我要写书评

暂无评论

Journal of Oceanology and Limnology 2020年第3期38卷 593-605页

作者： liu Xuehai liu yongzhi CAO Wei SUN Chengjun First Institute of Oceanography Ministry of Natural ResourcesQingdao 266061China Laboratory for Regional Oceanography and Numerical Modeling Pilot National Laboratory for Marine Science and Technology(Qingdao)Qingdao 266237China Key Laboratory of Marine Science and Numerical Modeling Ministry of Natural ResourcesQingdao 266061China

Using observations in an applied cruise of the submersible Jiaolong,water characteristics,geostrophic transport,and turbulent mixing in abyssal and hadal zones of the southern Yap Trench were *** spatial structures of deep water show that the abyssal water is cold,saline,and oxygen *** hadal water has very small changes in potential temperature and potential density,and a little decrease in salinity and obvious decrease in *** isotherm,isopycnal,and isohaline are depressed in abyss over the central *** turbulent mixing is enhanced in the near-bottom zone and the hadal water on the trench slope,especially at the steep slope,the dissipation rate and diffusivity is strong,which weakens the *** geostrophic flows move southward in the western region of the trench and northward in the eastern region,indicating cyclonic *** the central region of the trench,the water transport is^1.74 Sv *** the hadal zone,the northward and southward transports are *** analysis suggests that the abyssal water in the southern Yap Trench is from Lower Circumpolar Water(LCPW)and the hadal water seems to be of the isolated local water rather than LCPW.

关键词： Yap Trench abyssal and hadal the submersible Jiaolong diapycnal mixing geostrophic flows

基于改进ResNet101网络的齿轮缺陷检测

在线全文

学校读者我要写书评

暂无评论

组合机床与自动化加工技术 2024年第8期 145-148,153页

作者：包从望江伟刘永志肖钦兰吴娇六盘水师范学院矿业与机械工程学院六盘水553000 中国矿业大学机电工程学院徐州221116

针对视觉技术下的齿轮缺陷检测中,存在检测精度低、特征提取能力弱及检测模型不稳定等问题,提出了一种改进ResNet101网络的齿轮缺陷检测方法。首先,基于ResNet101网络,引入空洞卷积操作,在各个残差层中引入不同比例的膨胀系数,实现齿轮... 详细信息

针对视觉技术下的齿轮缺陷检测中,存在检测精度低、特征提取能力弱及检测模型不稳定等问题,提出了一种改进ResNet101网络的齿轮缺陷检测方法。首先,基于ResNet101网络,引入空洞卷积操作,在各个残差层中引入不同比例的膨胀系数,实现齿轮图像不同感受野下的特征提取;其次,在各个卷积模块间引入稠密连接操作,保留浅层特征信息,降低了模型训练过程中梯度消失的风险;最后,通过图像样本旋转操作,获得齿轮缺陷样本,通过准确率、召回率、ROC曲线、AUC等参数对所提方法的性能进行验证。实验结果表明,改进后的ResNet101能有效实现齿轮缺陷检测,同时具有更高的稳定性能,可用于齿轮生产过程中,产品质量的实时在线检测。

关键词：深度学习 ResNet101网络齿轮缺陷特征提取

人工智能时代网络主流意识形态安全风险治理对策研究

在线全文

学校读者我要写书评

暂无评论

燕山大学学报（哲学社会科学版） 2024年第5期25卷 64-72页

作者：刘永志史业达燕山大学马克思主义学院河北秦皇岛066004

人工智能作为新一轮产业革命的核心驱动力,以其核心技术算法、算力和大数据深刻改变了网络意识形态领域。人工智能技术的不断开发和应用,使得网络空间的意识形态斗争变得更加复杂化、隐匿化、多样化,网络主流意识形态安全建设任务更为... 详细信息

人工智能作为新一轮产业革命的核心驱动力,以其核心技术算法、算力和大数据深刻改变了网络意识形态领域。人工智能技术的不断开发和应用,使得网络空间的意识形态斗争变得更加复杂化、隐匿化、多样化,网络主流意识形态安全建设任务更为艰巨。人工智能技术与网络主流意识形态建设呈现双向互动的关系,人工智能技术赋能网络主流意识形态安全建设,网络主流意识形态安全建设为人工智能技术的互联网应用提供价值规范指引。人工智能技术异化效应与互联网“去中心化”价值取向耦合共振,使得网络主流意识形态呈现价值偏差、引领乏力、权威弱化的风险倾向。采取坚定马克思主义的指导地位、坚持培育社会主义价值共识、强化对人工智能的技术赋能、完善相应的制度体系等综合治理之策,有利于科学有效应对人工智能技术带来的意识形态安全风险,增强网络主流意识形态吸引力和凝聚力,营造风清气正的网络生态。

关键词：人工智能意识形态安全风险治理对策

维普期刊数据库博看期刊

基于最大均值差异的卷积神经网络故障诊断模型

在线全文

学校读者我要写书评

暂无评论

机电工程 2024年第3期41卷 445-454页

作者：包从望车守全刘永志陈俊张彩红六盘水师范学院矿业与机械工程学院贵州六盘水553000

针对工程场景中轴承故障数据采集困难,小样本下轴承故障诊断准确率较低且稳定性不高的问题,提出了一种小样本下滚动轴承故障的诊断方法,即基于最大均值差异(MMD)的卷积神经网络(CNN)诊断模型(方法)。首先,根据轴承故障机理,获取了滚动... 详细信息

针对工程场景中轴承故障数据采集困难,小样本下轴承故障诊断准确率较低且稳定性不高的问题,提出了一种小样本下滚动轴承故障的诊断方法,即基于最大均值差异(MMD)的卷积神经网络(CNN)诊断模型(方法)。首先,根据轴承故障机理,获取了滚动轴承故障的仿真信号,基于生成式对抗网络构建了仿真信号与少量真实样本间的对抗训练模型,得到了伪域样本,并将其扩充为训练数据集;其次,以交叉熵损失和最大均值差异(MMD)为卷积神经网络(CNN)的优化准则,引入了缩放因子,对网络进行了动态优化,根据测试结果选取缩放因子为0.05作为最优网络结构参数,构建了故障诊断的训练模型;最后,将结构均为1024个数据点的伪域样本和真实样本共同构成模型的训练集,对其进行了归一化处理,然后将其输入到构建的网络模型中,并以MMD作为约束,进行了卷积、池化操作,以实现特征提取的目的,经反向传播对模型进行了优化,实现了诊断模型参数的迭代更新目标。研究结果表明:基于MMD的CNN诊断模型(方法)对小样本下轴承的故障诊断识别精度有明显的改善,当样本数仅为16时,识别率可达95%以上,证明该方法在小样本下的轴承故障诊断中依然能获得较高的故障识别率。

关键词：滚动轴承故障诊断小样本生成式对抗网络卷积神经网络最大均值差异交叉熵损失