高级检索

基于改进神经网络和遗传算法的真空灭弧室优化设计

Optimal Design of Vacuum Interrupter Based on Improved Neural Network and Genetic Algorithm

  • 摘要: 提出一种基于人工神经网络和遗传算法相结合的电器电场均匀度优化设计方法,以12 kV真空灭弧室为研究对象,对其内部电场进行场影响因素分析与电极型面优化,以实现均匀场设计。建立了以悬浮屏蔽电极长度和端部型面曲率、动静触头电极端部曲率为输入变量,灭弧室内部电场均匀度为输出变量的改进神经网络模型结构;并以灭弧室内部电场均匀度为优化目标,采用遗传算法对灭弧室结构参数进行寻优,以提高灭弧室静态绝缘性能。

     

    Abstract: The optimal method for improving the electric field uniformity of vacuum interrupter based on the combination of artificial neural network and genetic algorithm is proposed,and taking a 12 kV vacuum interrupter as the research object.The internal electric field is analyzed by optimizing the influence factors of the structural parameters and the electrode contour.In order to achieve the uniform electric field design,the improved neural network model is proposed,considering the floating shield electrode length and end profile curvature,the curvature of the end of the movable and static electrodes as the input neurons and the uniformity of the electric field within the vacuum interrupter as the output variable.Furthermore,the uniformity of the electric field of the vacuum interrupter is determined as the optimal goal,the genetic algorithm is used to optimize the structural parameters of the vacuum interrupter,and the static insulation performance of the vacuum interrupter is improved.

     

/

返回文章
返回