|国家科技期刊平台
首页|期刊导航|电工技术学报|带铁心横磁-纵磁触头结构的优化设计

带铁心横磁-纵磁触头结构的优化设计OA北大核心CSTPCD

Optimal Design of Transverse Magnetic-Axial Magnetic Contact Structure with Iron Core

中文摘要英文摘要

触头作为真空灭弧室的核心元件,其灭弧能力会直接影响直流开断过程.该文通过有限元仿真软件建立了三种触头模型,计算了高频电流下触头产生的磁场分布.电流峰值时的触头间隙中心可产生约为 21 mT的纵向磁场,横向磁感应强度约为 71 mT,纵向磁场分布不均匀.为了进一步提高触头间隙磁场强度以及分布均匀度,在触头模型中增加铁心及改变铁心结构来提升触头产生的磁场强度,基于 BP神经网络与智能优化算法联合的方法对触头模型进行优化设计.优化结果显示:当触头的结构参数外触头非径向开槽夹角α为 31.8°、外触头内外半径差值Δ为 8.3 mm、内触头片半径r为 15.2 mm、外触头杯斜槽高度h为 16.1 mm、铁心结构开槽宽度d为 1.1 mm时,触头间隙中心产生了 74.99 mT的横向磁场;电流峰值时刻的触头间隙中心可产生 44.02 mT的纵向磁场,不均匀度从优化前的 7.44 减小到 4.67,均匀度得到了较大提升,提高了磁场对真空电弧的调控能力.

As the core component of vacuum interrupter,the contact's arc extinguishing ability will directly affect the DC breaking process.The existing contact models usually rely on transverse magnetic field or axial magnetic field to regulate the vacuum arc and help the vacuum circuit breaker to break.In this paper,a type of contact which can generate both transverse magnetic field and axial magnetic field is designed.The main design idea of the contact is that the vacuum arc generated on the surface of the transverse magnetic contact rotates and diffuses into the axial magnetic field under the action of electromagnetic force,and the arc changes from the concentrated state to the diffused state under the action of the axial magnetic field.The vacuum arc in the diffused state can greatly reduce the ablation on the contact surface and improve the service life of the contact. Firstly,three contact models are established by finite element simulation software,and the magnetic field distribution generated by the contact under high frequency current is calculated.At the peak of current,the contact gap center can generate an axial magnetic field of about 21 mT,the transverse magnetic field intensity is about 71 mT,and the axial magnetic field distribution is uneven. Secondly,due to the increase of iron core structure,the original divergent magnetic field can be bound into a certain space,and the magnetic field intensity in the contact gap can be enhanced.In this paper,an iron core structure is added between the inner and outer contacts,and the iron core structure is slotted.The simulation structure shows that after adding the iron core,the transverse magnetic field intensity increases by 1.22 mT,the maximum axial magnetic field intensity increases from 21.80 mT to 26.53 mT,the magnetic field intensity increases by 4.73 mT,the axial magnetic field distribution unevenness decreases from 8.90 to 7.39,and the magnetic field distribution becomes more uniform.Adding grooving treatment in the iron core structure increases the transverse magnetic field intensity in the contact gap by 0.25 mT,and the maximum axial magnetic field intensity also slightly increases by 0.49 mT,but the uneven distribution of axial magnetic field increases from 7.39 to 7.44,which is not conducive to the diffusion of vacuum arc. Finally,in order to further improve the strength and distribution uniformity of the magnetic field in the contact gap,this paper proposes a joint optimization method based on BP neural network and multi-objective intelligent optimization algorithm to optimize the contact model,establishes a BP neural network model with α,Δ,r,h and d as inputs and M,η and N as outputs,and optimizes the contact model through multi-objective intelligent optimization algorithm.The optimization results show that when the contact structure parameters α is 31.8°,Δ is 8.3 mm,r is 15.2 mm,h is 16.1 mm,and d is 1.1 mm,a transverse magnetic field of 74.99 mT is generated in the contact gap center.The axial magnetic field of 44.02 mT can be generated at the contact gap center of the peak current,and the unevenness is reduced from 7.44 before optimization to 4.67,which greatly improves the uniformity and improves the ability of magnetic field to regulate vacuum arc.

丁璨;李江;袁召;王周琳

三峡大学电气与新能源学院 宜昌 443002||电工材料电气绝缘全国重点实验室(西安交通大学) 西安 710049三峡大学电气与新能源学院 宜昌 443002强电磁技术全国重点实验室(华中科技大学) 武汉 430074

动力与电气工程

真空灭弧室触头横向磁场纵向磁场算法优化BP神经网络

Vacuum interrupter contacttransverse magnetic fieldaxial magnetic fieldalgorithm optimizationBP neural network

《电工技术学报》 2024 (011)

3499-3509 / 11

国家自然科学基金面上项目资助(52177143).

10.19595/j.cnki.1000-6753.tces.230474

评论