磁绝缘线振荡器的自动优化设计  

Self-optimization design of magnetically insulated transmission line oscillator

作　　者：孙会芳[1] 李瀚宇[1] 姜幼明[1] 董烨[1] 董志伟[1] 周海京[1] 

机构地区：[1]北京应用物理与计算数学研究所,北京100094

出　　处：《强激光与粒子束》2014年第4期17-21,共5页High Power Laser and Particle Beams

基　　金：中国工程物理研究院科学技术发展基金项目(2011B0402068;2012B0402064);国家自然科学基金项目(11305015)

摘　　要：为克服全电磁粒子模拟(PIC)程序不利于优化设计的弱点,提高高功率微波器件的优化设计水平,将遗传算法与全电磁粒子模拟算法有机融合,研制出二维全电磁粒子模拟并行优化程序。据此对高功率微波源器件——两个波段的磁绝缘线振荡器(MILO):C-MILO和L-MILO进行优化设计。在输入功率不变的条件下,原C-MILO效率为10.8%,经优化后效率为15.4%;原L-MILO效率为12.6%,经优化后效率为17.7%。由此得出,两类MILO模型经优化后在输入功率基本不变的情况下输出功率和效率都有很大程度的提高,且模型几何参数合理,物理图像正确。A paralleled 2-dimension optimization code syncretized Genetic Algorithm and PIC code is developed to improve the level of design of high power microwave(HPM)sources.Both C-MILO and L-MILO are optimized using this optimization code.By optimization design,the efficiency of C-MILO rises to 15.4%from 10.8%and that of L-MILO rises from 12.6%to 17. 7% while the input power keeps the same.It is received that both the output power and the efficiency of the two MILOS are improved considerably as their input power are changeless.All optimized models have reasonable geometry parameters and exact simulating spectra.

关 键 词：遗传算法 PIC程序 高功率微波器件 磁绝缘线振荡器 优化设计 

分 类 号：TN753.4[电子电信—电路与系统]