检索结果-内蒙古大学图书馆

On the importance of proper kernel normalization procedure in nonlocal integral continuum modeling of nanobeams

ZAMM-ZEITSCHRIFT FUR ANGEWANDTE MATHEMATIK UND MECHANIK 2021年第10期101卷

作者： Anjomshoa, Amin Hassani, Behrooz Ferdowsi Univ Mashhad Dept Mech Engn Mashhad 917751111 Razavi Khorasan Iran

Increasing the bending analysis accuracy of nanobeams employing the nonlocal integral (NI) models is the subject of this paper. It is shown in this study that if the kernel of the NI model is not properly normalized, the natural boundary conditions will be violated and the deflections will be noticeably overestimated. Two kernel normalization procedures, borrowed from other disciplines, are rigorously examined and their influence on the nanobeam deflections is studied in detail. It is shown that one of these procedures, which has recently been used for the analysis of nanobeams, causes the NI model to show both softening and hardening behaviors with unexpected convergence to the classical continuum model for very large nonlocal parameters. This surprising behavior, makes the NI model incapable of being calibrated in some cases and may lead to two distinct calibrated nonlocal parameters which is confusing. This is confirmed here via a comparison with atomistic models. The other normalization scheme shows a consistent softening behavior without such shortcomings and therefore is proposed here for bending analysis of nanobeams. Effect of kernel function distribution, as a representative of interatomic cohesive forces, and edge support condition on the maximum deflection of normalized nonlocal nanobeam model is also thoroughly investigated. Moreover, the effect of surface energy is also investigated via a combined model.

关键词： kernel function type kernel normalization nanobeam bending analysis natural boundary conditions nonlocal integral theory

来源：评论

学校读者我要写书评

暂无评论

Improving the Accuracy of the SVM Classification using the Parzen Classifier 7

Improving the Accuracy of the SVM Classification using the P...

引用

7th Mediterranean Conference on Embedded Computing (MECO)

作者： Demidova, Liliya Egin, Maksim Ryazan State Radio Engn Univ Gagarin Str 59-1 Ryazan 390005 Russia

ISBN: (纸本)9781538656839

In this paper the data classification technique, implying the consistent application of the SVM and Parzen classifiers, has been suggested. The Parzen classifier allows improving the accuracy of the data classification on the basis of the SVM classifier. The Parzen classifier is applied for data of subareas near the hyperplane dividing the classes. These subareas are determined experimentally for the concrete dataset. The recommendations on the use of the Parzen classifier near the hyperplane dividing the classes have been provided. The examples of application of the offered technique for the solving the data classification problem have been given.

关键词： SVM classifier support vectors kernel function type kernel function parameters regularization parameter Parzen classifier

来源：评论

学校读者我要写书评

暂无评论

A Novel SVM-kNN Technique for Data Classification 6

A Novel SVM-kNN Technique for Data Classification

引用

6th Mediterranean Conference on Embedded Computing (MECO)

作者： Demidova, Liliya Sokolova, Yulia Ryazan State Radio Engn Univ Gagarin Str 59-1 Ryazan 390005 Russia

ISBN: (纸本)9781509067428

In this paper the data classification technique, implying the consistent application of the SVM and kNN classifiers in the respective determined experimentally subareas of the characteristics space has been suggested. The recommendations on the use of the kNN classifier near the hyperplane dividing the classes have been provided. The examples of application of the offered technique for the problem of data classification have been given.

关键词： SVM classifier support vectors kernel function type kernel function parameters regularization parameter kNN classifier classification technique

来源：评论

学校读者我要写书评

暂无评论

Two-Level Intellectual Classifier Based on the SVM Algorithm 6

Two-Level Intellectual Classifier Based on the SVM Algorithm

引用

6th Mediterranean Conference on Embedded Computing (MECO)

作者： Demidova, Liliya Sokolova, Yulia Ryazan State Radio Engn Univ Gagarin Str 59-1 Ryazan 390005 Russia

ISBN: (纸本)9781509067428

In this paper the two-level intellectual classifier based on the SVM algorithm has been offered. This classifier works as the group of the SVM classifiers at the first level and as the final SVM classifier on the base of the modified PSO algorithm at the second level. Herewith, we use the original experimental dataset to develop the different SVM classifiers at the first level, and the new dataset, containing only supports vectors of the SVM classifiers of the first level, to develop the final SVM classifier on the base of the modified PSO algorithm at the second level. The results of experimental studies confirm the efficiency of the offered two-level intellectual classifier.

关键词： SVM classifier support vectors kernel function type kernel function parameters regularization parameter particle swarm optimization algorithm

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：