A detection method of Edge Coherent Mode based on improved SSD

被引：11

作者：

Zeng, Fulin ^{[1
]}

Liu, Ying ^{[1
]}

Ye, Yang ^{[2
]}

Zhou, Jijun ^{[1
]}

Liu, Xiaotao ^{[1
]}

机构：

[1] Shenzhen Univ, SZU CASIPP Joint Lab Appl Plasma, Shenzhen 518060, Peoples R China

[2] Hefei Comprehens Natl Sci Ctr, Inst Energy, Hefei 230031, Peoples R China

来源：

FUSION ENGINEERING AND DESIGN | 2022年 / 179卷

基金：

中国国家自然科学基金;

关键词：

EAST; Edge coherent mode; Deep learning; Object detection;

D O I：

10.1016/j.fusengdes.2022.113141

中图分类号：

TL [原子能技术]; O571 [原子核物理学];

学科分类号：

0827 ; 082701 ;

摘要：

It has become a trend to deeply analyze and mine diagnostic data using machine learning methods. Single-shot MultiBox detector (SSD) is an excellent target detection network in deep learning. In this paper, two improvements for SSD are proposed to accurately detect the edge coherent mode in the cross-power spectrum. One is adjusting the positive-negative sample ratios to balance the training dataset, and the other is importing the flooding loss to mitigate overfitting. After testing, the accuracy of the improved SSD reaches 92.12%. The result shows that the performance of the improved SSD for ECM detecting is better than that of the typical SSD, which can be applied to subsequent intelligent data analysis.

引用

页数：7

共 16 条

[1]

[Anonymous], 2015, 3 INT C LEARN REPR I

[2] Towards an automatic filament detector with a Faster R-CNN on MAST-U [J].

Cannas, B. ;

Carcangiu, S. ;

Fanni, A. ;

Farley, T. ;

Militello, F. ;

Montisci, A. ;

Pisano, F. ;

Sias, G. ;

Walkden, N. .

FUSION ENGINEERING AND DESIGN, 2019, 146 :374-377

[3] The PASCAL Visual Object Classes Challenge: A Retrospective [J].

Everingham, Mark ;

Eslami, S. M. Ali ;

Van Gool, Luc ;

Williams, Christopher K. I. ;

Winn, John ;

Zisserman, Andrew .

INTERNATIONAL JOURNAL OF COMPUTER VISION, 2015, 111 (01) :98-136

[4]

Hoiem D, 2012, LECT NOTES COMPUT SC, V7574, P340, DOI 10.1007/978-3-642-33712-3_25

[5]

Kingma D. P., 2015, INT C LEARN REPR

[6] Tokamak visible image sequence recognition using nonlocal spatio-temporal CNN for attention needed area localization* [J].

Kwon, Giil ;

Wi, Hanmin ;

Hong, Jaesic .

FUSION ENGINEERING AND DESIGN, 2021, 168

[7]

Li Jian-Gang, 2016, Wuli, V45, P88, DOI 10.7693/wl20160203

[8] SSD: Single Shot MultiBox Detector [J].

Liu, Wei ;

Anguelov, Dragomir ;

Erhan, Dumitru ;

Szegedy, Christian ;

Reed, Scott ;

Fu, Cheng-Yang ;

Berg, Alexander C. .

COMPUTER VISION - ECCV 2016, PT I, 2016, 9905 :21-37

[9] An intelligent identification method of the edge coherent mode in EAST [J].

Liu, Ying ;

Yang, Yuqian ;

Huang, Jianjun ;

Yang, Ye ;

Wang, Fan .

FUSION ENGINEERING AND DESIGN, 2019, 145 :6-10

[10] Classification of tokamak plasma confinement states with convolutional recurrent neural networks [J].

Matos, F. ;

Menkovski, V. ;

Felici, F. ;

Pau, A. ;

Jenko, F. .

NUCLEAR FUSION, 2020, 60 (03)

← 1 2 →