摘要
针对传统天线设计方法计算代价较大的缺陷,该文构建基于反向传播神经网络(BPNN)的新型天线代理模型。为解决BPNN训练易陷入局部最优的问题,采用粒子群优化(PSO)算法来改善神经网络初始结构参数,进而构建PSO-BPNN天线代理模型,并基于该模型提出多参数天线结构的快速多目标设计方法。设计实例表明,该方法在预测精度以及计算代价等方面优于现有的常用天线设计方法。所提方法对处理复杂高维参数空间天线设计问题具有实用价值。
Focusing on the problem of reducing the large computation cost of traditional antenna design methods, a new surrogate model based on Back Propagation Neural Networks(BPNN) is constructed. In order to solve the problem of easily falling into local optimum in BPNN, a PSO-BPNN surrogate model is developed by improving initial structural parameters of neural networks and applied to fast multi-objective optimization design of multi-parameter antenna structures. The design results show that the proposed PSO-BPNN outperforms other existing antenna surrogate models in terms of prediction accuracy and prediction speed. The proposed method is of value in dealing with complex antenna designs with high-dimensional parameter space.
引文
[1]MOHAMMED H J,ABDULLAH,A S,ALI R S,et al Design of a uniplanar printed triple band-rejected ultrawideband antenna using particle swarm optimisation and the firefly algorithm[J].IET Microwaves,Antennas&Propagation,2016,10(1):31-37.doi:10.1049/ietmap.2014.0736.
[2]CHOI K,JANG D,KANG S,et al.Hybrid algorithm combing genetic algorithm with evolution strategy for antenna design[J].IEEE Transactions on Magnetics,201652(3):7209004.doi:10.1109/TMAG.2015.2486043.
[3]GOUDOS S K,KALIALAKIS C,and MITTRA R Evolutionary algorithms applied to antennas and propagation:A review of state of the art[J].International Journal of Antennas and Propagation,2016,2016(4):1-12.doi:10.1155/2016/1010459.
[4]KOZIEL S and OGURTSOY S.Multi-objective design of antennas using variable-fidelity simulations and surrogate models[J].IEEE Transactions on Antennas and Propagation,2013,61(12):5931-5939.doi:10.1109/TAP.2013.2283599.
[5]陈晓辉,裴进明,郭欣欣,等.一种基于多维均匀采样与Kriging模型的天线快速优化方法[J].电子与信息学报,2014,36(12):3021-3026.doi:10.3724/SP.J.1146.2013.01826.CHEN Xiaohui,PEI Jinming,GUO Xinxin,et al.An efficient antenna optimization method based on kriging model and multidimensional uniform sampling[J].Journal of Electronics&Information Technology,2014,36(12):3021-3026.doi:10.3724/SP.J.1146.2013.01826.
[6]DONG Jian,LI Qianqian,and DENG Lianwen.Fast multiobjective optimization of multi-parameter antenna structures based on improved MOEA/D with surrogateassisted model[J].AEUE-International Journal of Electronics and Communications,2017,72:192-199.doi:10.1016/j.aeue.2016.12.007.
[7]LIU Bo,ALIAKBARIAN H,MA Zhongkun,et al.An efficient method for antenna design optimization based on evolutionary computation and machine learning techniques[J].IEEE Transactions on Antennas and Propagation,2014,62(1):7-18.doi:10.1109/TAP.2013.2283605.
[8]JACOBS J P.Efficient resonant frequency modeling for dual-band microstrip antennas by Gaussian process regression[J].IEEE Antennas and Wireless Propagation Letters,2015,14:337-341.doi:10.1109/LAWP.2014.2362937.
[9]CHEN Linglu,LIAO Cheng,LIN Wenbin,et al.Hybridsurrogate-model-based efficient global optimization for highdimensional antenna design[J].Progress in Electromagnetics Research,2012,124(8):85-100.doi:10.2528/PIER11121203.
[10]MASSA A,OLIVERI G,SALUCCI M,et al.Learning-byexamples techniques as applied to electromagnetics[J].Journal of Electromagnetic Waves and Applications,2017,32(4):516-541.doi:10.1080/09205071.2017.1402713.
[11]焦李成,杨淑媛,刘芳,等.神经网络七十年:回顾与展望[J].计算机学报,2016,39(8):1697-1716.doi:10.11897/SP.J.1016.2016.01697.JIAO Licheng,YANG Shuyuan,LIU Fang,et al.Seventy years beyond neural networks:retrospect and prospect[J].Chinese Journal of Computers,2016,39(8):1697-1716.doi:10.11897/SP.J.1016.2016.01697.
[12]公茂果,焦李成,杨咚咚,等.进化多目标优化算法研究[J].软件学报,2009,20(2):271-289.doi:10.3724/SP.J.1001.2009.03483.GONG Maoguo,JIAO Licheng,YANG Dongdong,et al.Research on evolutionary multi-objective optimization algorithms[J].Journal of Software,2009,20(2):271-289.doi:10.3724/SP.J.1001.2009.03483.
[13]RUMELHART D E,HINTON G E,and WILLIAMS R J.Learning representations by back-propagating errors[J].Nature,1986,323(9):533-536.doi:10.1038/323533a0.
[14]KOLMOGOROV A N.On the representation of continuous functions of several variables by superposition of continuous functions of one variable and addition[J].Doklady Akademii Nauk SSSR,1957,114(5):953-956.doi:10.1007/978-94-011-3030-1_56.
[15]STEIN M.Large sample properties of simulations using Latin hypercube sampling[J].Technometrics,1987,29(2):143-151.doi:10.1080/00401706.1987.10488205.
[16]KENNEDY J and EBERHART R C.Particle swarm optimization[C].Proceedings of IEEE International Conference on Neural Networks,Perth,Australia,1995,4:1942-1948.doi:10.1109/icnn.1995.488968.
[17]COELLO C A C,PULIDO G T,and LECHUGA M S.Handling multiple objectives with particle swarm optimization[J].IEEE Transactions on Evolutionary Computation,2004,8(3):256-279.doi:10.1109/TEVC.2004.826067.
[18]DONG Jian,YU Xiaping,and HU Guoqiang.Design of a compact quad-band slot antenna for integrated mobile devices[J].International Journal of Antennas and Propagation,2016,2016:1-9.doi:10.1155/2016/3717681.
[19]ANURDHA,PATNAIK A,and SINHA S N.Design of custom-made fractal multi-band antennas using ANN-PSO[J].IEEE Antennas&Propagation Magazine,2011,53(4):94-101.doi:10.1109/MAP.2011.6097296.
[20]ROBINSON J and RAHMAT-SAMMI Y.Particle swarm optimization in electromagnetics[J].IEEE Transactions on Antennas and Propagation,2004,52(2):397-407.doi:10.1109/TAP.2004.823969.
[21]JIN Nanbo and RAHMAT-SAMMI Y.Advances in particle swarm optimization for antenna designs:Real-number,b i n a r y,s i n g l e-o b j e c t i v e a n d m u l t i o b j e c t i v e implementations[J].IEEE Transactions on Antennas and Propagation,2007,55(3):556-567.doi:10.1109/TAP.2007.891552.