基于粒子模拟和并行遗传算法的高功率微波源优化设计
Optimal design of high-power microwave source based on particle simulation and genetic algorithms
-
摘要: 提出了基于粒子模拟和并行遗传算法的高功率微波源优化设计方法,以全电磁粒子模拟软件UNIPIC模拟的高功率微波器件输出功率作为适应度函数,采用浮点数编码的遗传算法对高功率微波源器件进行优化。采用该算法,对相对论返波管的布拉格反射器位置以及高度进行了浮点数编码,然后在巨型机上进行参数的全局优化,获得了该返波管布拉格反射器的全局最优参数。Abstract: Optimal design method of high-power microwave source based on particle simulation and parallel genetic algorithm is presented in this paper. The output power of the high-power microwave device, simulated by the fully electromagnetic particle simulation code UNIPIC, is given as a fitness function, and the float-encoding genetic algorithm is used to optimize the high-power microwave device. Using this method, we encode the position and height of the Bragg reflector of the relativistic backward wave oscillator (RBWO), and optimize the parameters for a massively parallel processor. Simulation results demonstrate that we can obtain the global optimal parameters of the Bragg reflector of the RBWO.
-
-
计量
- 文章访问数: 548
- HTML全文浏览数: 209
- PDF下载数: 1
- 施引文献: 0
首页
登录
注册


下载: