电工技术学报  2022, Vol. 37 Issue (23): 6074-6082    DOI: 10.19595/j.cnki.1000-6753.tces.220369
电工理论 |
基于NSGA-II和BP神经网络的杯状纵磁触头结构优化设计
丁璨1, 李江1, 袁召2, 陈立学2, 刘黎明2
1.三峡大学电气与新能源学院 宜昌 443002;
2.华中科技大学电气与电子工程学院强电磁工程与新技术国家重点实验室 武汉 430074
Structural Optimization Design of Cup-Shaped Longitudinal Magnetic Contact Based on NSGA-II and BP Neural Network
Ding Can1, Li Jiang1, Yuan Zhao2, Chen Lixue2, Liu Liming2
1. College of Electrical Engineering & New Energy China Three Gorges University Yichang 443002 China;
2. State Key Laboratory of Advanced Electromagnetic Engineering and Technology School of Electrical and Electronic Engineering Huazhong University of Science and Technology Wuhan 430074 China
全文: PDF (6461 KB)   HTML
输出: BibTeX | EndNote (RIS)      
摘要 在直流开断工况下,现有常用工频触头需要开断高频反向电流,但其在高频电流下的电弧调控性能还需要进一步研究。该文以工频常用杯状纵磁触头为例,首先对其在1 000Hz电流下进行磁场仿真,发现电流峰值的触头间隙中心可产生3.17mT/kA的纵向磁场,磁场分布极不均匀;然后为了提高磁场强度最大值以及分布均匀度,提出一种基于BP神经网络与遗传算法相结合的方法对触头结构进行优化设计,建立以触头开槽宽度、开槽长度、径向开槽旋转角度、触头杯斜槽高度和斜槽倾斜度为输入,触头间隙中心电流峰值磁场强度最大值和磁场分布不均匀度为输出的BP神经网络模型;最后通过NSGA-II算法对杯状纵磁触头的结构参数进行优化。优化的结果表明:当杯状纵磁触头的参数L1为2.9mm、L2为18.0mm、θ1为20.0°、H为17.5mm和θ2为26.0°时,电流峰值的触头间隙中心可产生4.34mT/kA的纵向磁场,不均匀度从6.89减小到3.39,均匀度得到了较大提升,从而可以提高纵向磁场对真空电弧的调控能力。
服务
把本文推荐给朋友
加入我的书架
加入引用管理器
E-mail Alert
RSS
作者相关文章
丁璨
李江
袁召
陈立学
刘黎明
关键词 真空灭弧室触头纵向磁场NSGA-IIBP神经网络    
Abstract:Under the DC breaking condition, the existing common power frequency contacts need to break the high-frequency reverse current, but their arc regulation performance under high-frequency current needs further study. In this paper, the cup-shaped longitudinal magnetic contact commonly used in power frequency is taken as an example. Firstly, the magnetic field simulation of the cup-shaped longitudinal magnetic contact is carried out under the current of 1 000Hz. It is found that the center of the contact gap with the peak current can generate a longitudinal magnetic field of 3.17mT/kA, and the magnetic field distribution is extremely uneven. Then, in order to improve the maximum magnetic field strength and distribution uniformity, a method based on the combination of BP neural network and genetic algorithm is proposed to optimize the contact structure, and a BP neural network model is established, which takes the contact slot width, slot length, radial slot rotation angle, contact cup chute height and chute inclination as inputs, and takes the maximum magnetic field strength and distribution unevenness of the peak current in the center of the contact gap as outputs. Finally, the structural parameters of the cup-shaped longitudinal magnetic contact are optimized by NSGA-II algorithm. The optimization results show that when the parameters L1 is 2.9mm, L2 is 18.0 mm, θ1 is 20.0°, H is 17.5mm and θ2 is 26.0°, the center of the contact gap with peak current can generate a longitudinal magnetic field of 4.34mT/kA, and the unevenness is reduced from 6.89 to 3.39, which greatly improves the uniformity, thus improving the ability of the longitudinal magnetic field to regulate the vacuum arc.
Key wordsVacuum interrupter contact    longitudinal magnetic field    NSGA-II    BP neural network   
收稿日期: 2022-03-12     
PACS: TM561  
基金资助:国家自然科学基金面上项目资助(52177143)
通讯作者: 袁 召 男,1985年生,博士(后),副研究员,研究方向为电气设备设计、仿真模拟与优化。E-mail:hustyuanzhao@qq.com   
作者简介: 丁 璨 男,1982年生,博士,讲师,研究方向为高压电器设计、模拟及实验。E-mail:dingcan@ctgu.edu.cn
引用本文:   
丁璨, 李江, 袁召, 陈立学, 刘黎明. 基于NSGA-II和BP神经网络的杯状纵磁触头结构优化设计[J]. 电工技术学报, 2022, 37(23): 6074-6082. Ding Can, Li Jiang, Yuan Zhao, Chen Lixue, Liu Liming. Structural Optimization Design of Cup-Shaped Longitudinal Magnetic Contact Based on NSGA-II and BP Neural Network. Transactions of China Electrotechnical Society, 2022, 37(23): 6074-6082.
链接本文:  
https://dgjsxb.ces-transaction.com/CN/10.19595/j.cnki.1000-6753.tces.220369          https://dgjsxb.ces-transaction.com/CN/Y2022/V37/I23/6074