Construction of Permittivity-Graded Insulator Using Topology-Optimized Lattice Structure

IEEE TRANSACTIONS ON DIELECTRICS AND ELECTRICAL INSULATION(2024)

引用 0|浏览8
暂无评分
摘要
Dielectrically graded insulation (DGI) is an effective means to mitigate the electric field ( ${E}$ -field) distribution of insulators. The industrial application of DGI remains at an exploratory stage due to the challenge of accurately constructing gradient dielectric properties by an efficient approach. Periodic lattice material (PLM) is a cellular material with periodically arranged unit cells, which can be used to construct graded material systems. In this study, the applicability of PLM in DGI is investigated. The PLM is composed of lattice unit cells with cylinder trusses and solid fillings, and the spatial distribution of permittivity can be modified by changing the permittivity of each unit cell in different regions. First, a quantitative prediction model of PLM's permittivity from geometrical and material parameters (truss radius, truss permittivity, and filling permittivity) was built using finite element simulation and a feedforward neural network (FNN). The prediction error to the experimental data is less than 4.93%. Second, a construction method of lattice-graded insulator was developed and was used on a truncated cone insulator to reduce the ${E}$ -field concentration at its triple junctions. Finally, a fabrication strategy of the lattice-graded insulator combining 3-D printing and vacuum casting was proposed. The ac flashover voltage (FOV) tests show that the lattice-graded insulator exhibits a maximum FOV improvement of 14.6%, 57.8%, and 48.3% in 0.1 MPa air, C4F7N/CO2, and SF6, respectively. It is thus expected that PLM is promising for accurate and efficient DGI insulations, having the potential to be applied in actual insulation devices.
更多
查看译文
关键词
Lattices,Insulators,Permittivity,Insulation,Optimization,Finite element analysis,Fabrication,3-D printing,dielectrically graded insulation (DGI),lattice material,solid insulation,surface flashover,topology optimization
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要