Publishing Service

Polishing & Checking

Journal of Zhejiang University SCIENCE A

ISSN 1673-565X(Print), 1862-1775(Online), Monthly

2022 Vol.23 No.4 P.329-334

Eccentric actuator driven by stacked electrohydrodynamic pumps

Ze-bing MAO, Yota ASAI, Ardi WIRANATA, De-qing KONG, Jia MAN

Smart Materials Lab, Department of Engineering Science and Mechanics, Shibaura Institute of Technology, Tokyo 135-8548, Japan; Department of Mechanical and Industrial Engineering, Faculty of Engineering, University of Gadjah Mada, Yogyakarta 55281, Indonesia; Muroran Institute of Technology, Muroran, Hokkaido, Japan; Key Laboratory of High Efficiency and Clean Mechanical Manufacture of MOE, School of Mechanical Engineering, Shandong University, Jinan 250061, China; Introduction

zebingv5@shibaura-it.ac.jp

Abstract: In this study, we designed, fabricated, and characterized single-layer and stacked EHD pumps and applied them to drive an eccentric actuator. The proposed single EHD pumps were composed of transparent base and covers, and conductive copper electrodes. To boost the output power, we also developed stacked EHD pumps, made of four layers of EHD pumps. We used the Novec HFE 7300 as the pumping medium. The pumps and eccentric actuators were realized by using the two methods of digital fabrication and replication mold, respectively.

Key words: Eccentric actuator; Stacked electrohydrodynamic pump

Chinese Summary  <23> 一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

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
1江苏大学计算机科学与通信工程学院，中国镇江市，212013
2江苏大学工业网络空间安全技术江苏省重点实验室，中国镇江市，212013
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。

关键词组：缺陷预测；特征选择；迁移学习；度量补偿


Share this article to： More

Go to Contents

References:

<Show All>

Open peer comments: Debate/Discuss/Question/Opinion

<1>

Please provide your name, email address and a comment





DOI:

10.1631/jzus.A2100468

CLC number:

Download Full Text:

Click Here

Downloaded:

1296

Download summary:

<Click Here> 

Downloaded:

306

Clicked:

1085

Cited:

0

On-line Access:

2022-05-10

Received:

2021-09-21

Revision Accepted:

2021-12-22

Crosschecked:

2022-05-11

Journal of Zhejiang University-SCIENCE, 38 Zheda Road, Hangzhou 310027, China
Tel: +86-571-87952276; Fax: +86-571-87952331; E-mail: jzus@zju.edu.cn
Copyright © 2000~ Journal of Zhejiang University-SCIENCE