基于引力柯西多目标猫群算法的慢波结构优化  

Slow-wave structure optimization based on multi-objective Cauchy mutated cat swarm optimization with gravitational search operator

作　　者：赵东明 夏克文 刘慧娟 石晓旭 ZHAO Dong-ming;XIA Ke-wen;LIU Hui-juan;SHI Xiao-xu(School of Electronic and Information Engineering,Hebei University of Technology,Tianjin 300401,China;Film and Television Technology Department,Tianjin Broadcasting TV and Film Institute,Tianjin 300112,China;Hangzhou Zhijiang Switchgear Limited Company,Hangzhou 311200,China)

机构地区：[1]河北工业大学电子信息工程学院,天津300401 [2]杭申集团杭州之江开关股份有限公司,浙江杭州311200 [3]天津广播影视职业学院影视技术系,天津300112

出　　处：《计算机工程与设计》2019年第2期418-424,共7页Computer Engineering and Design

基　　金：河北省自然科学基金项目(E2016202341);河北省高等学校科学技术研究基金项目(BJ2014013)

摘　　要：为解决空间行波管慢波结构的电子效率和相移量多目标优化问题,提出引力柯西多目标猫群算法(GS-MOCMCSO),以电子效率提升和相移量降低为目标函数,结合非线性频谱模型1-D CHRISTINE,持续进行迭代优化。实验结果表明,GS-MOCMCSO算法收敛速度快、精度高,电子效率相比额定值提升10.2%,相移量相比初始值降低23.5°。与经典多目标猫群算法(MOCSO)相比,GS-MOCMCSO算法的最佳电子效率更高(40.2%对38.5%),相移量相近,带宽得到扩展(2GHz对1.5GHz),更适合于面向电子效率和相位失真的多目标优化。To solve the multi-objective optimization problem of slow-wave structure in space traveling-wave tube,a method based on multi-objective Cauchy mutated cat swarm optimization with gravitational search operator(GS-MOCMCSO)was proposed,electron beam efficiency and phase shift were used as the two objective functions,and 1-D CHRISTINE code was introduced to obtain the output value,the best pitch distribution by GS-MOCMCSO was calculated through continuous optimization.Experimental results demonstrate that the electronic beam efficiency on the best pitch distribution obtained using GS-MOCMCSO reaches 40.2%,promoted by 10.2%compared with rated value 30%,and it is higher than that of classical multi-objective cat swarm optimization(MOCSO,38.5%).The phase shift is only 36.5°,promoted by 23.5°compared with the value(60°)under constant pitch,bandwidth is extended to 2 GHz and is better than that of MOCSO(1.5 GHz).Thus,the proposed GS-MOCMCSO is suitable for the slow-wave structure optimization of STWT,and the GS-MOCMCSO is superior to MOCSO for the MOP of beam efficiency and phase shift.

关 键 词：柯西猫群算法 引力搜索算子 慢波结构 多目标优化 电子效率 相移量 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术]