Abstract：An improved genetic algorithm, called adaptive dual-subpopulation genetic algorithm (ADSGA) is proposed to overcome the drawbacks of genetic algorithm(GA), such as premature convergence and deficiency of local search etc. Based on GA, ADSGA introduces the dual-subpopulation and adaptive termination condition based on the number of generation. During the manipulation of ADSGA, the best available individual is firstly obtained. Then the symmetric transformation of the currently population are operated via the best individual, which results in subpopulation. Thereafter environmental selection is applied to keep the elites of the original population and its rival to create new population for the next generation. By means of benchmark test function, ADSGA shows superior in convergence rate and robustness compared with GA.In the end, the proposed algorithm is applied to optimize superconducting magenetic energy storage system(SMES). The satisfying result proves the merit of ADSGA for electromagnetic devices optimization.
徐斌, 姚缨英. 自适应对偶种群遗传算法及其在电磁场优化设计中的应用[J]. 电工技术学报, 2013, 28(3): 183-187.
Xu Bin, Yao Yingying. Adaptive Dual-Subpopulation Genetic Algorithm and Its Application for Electromagnetic Devices Optimization. Transactions of China Electrotechnical Society, 2013, 28(3): 183-187.
[1] Chen Tanggong, Wang Youhua, Liu Zibin, et al.Cyclic shift genetic algorithm applied to design optimization of electromagnetic devices[C]. Proceedings of the International Conference on Electrical Machines and Systems, 2008: 189-192. [2] Sofiene Saidi, Jaleleddine Slama. Effect of genetic algorithm parameters on convergence of the electromagnetic inverse method[C]. Proceedings of the 8th International Multi-Conference on Systems, Signals & Devices, 2011: 1-5. [3] Li Junhua, Li Ming, Li Junhua. Genetic algorithm with dual species[C]. Proceedings of the International Conference on Automation and Logistics, 2008: 2572-2575. [4] RahilaPatel, Raghuwanshi M M, Anil jaiswal modifying genetic algorithm with species and sexual selection by using k-means algorithm[C]. Proceedings of the IEEE International Advance Comnputing Conference, 2009: 114-119. [5] Deng Xianghui. Application of adaptive genetic algorithm in inversion analysis of permeability coefficients[C]. Proceedings of the Second International Conference on Genetic and Evolutionary Computing, 2008: 61-65. [6] Chen Tanggong, Wang Youhua, Pang Lingling, et al. Adaptive population disappearance genetic algorithm for electromagnetic devices optimization automation congress[C]. Proceedings of the the 6th International Conference on Electromagnetic Field Problems and Applications, 2008: 1-4. [7] Guillaume Crevecoeur, Peter Sergeant, Luc Dupre, et al. A two-level genetic algorithm for electromagnetic optimization[J]. IEEE Transactions on Magnetics, 2010, 46(7): 2585-2595. [8] 杨庆新, 安金龙, 马振平, 等.基于最小二乘支持向量机和自适应模拟退火算法的电磁场逆问题全局优化方法[J]. 电工技术学报, 2008, 23(11): 1-7. Yang Qinxin, An Jinlong, Ma Zhenping, et al. A global vector M optimization algorithm ba-sed on least squares support achines and adaptive simulated annealing algorithm for in-verse electromagnetic problem[J]. Transactions of China Electrotechnical Society, 2008, 23(11): 1-7. [9] 谢德馨, 杨仕友. 工程电磁场数值分析与综合[M].北京: 机械工业出版社, 2009. [10] TEAM optimization benchmark problem 22 [OL]. www. igte.tugraz.at. [11] 张慧. 电磁场逆问题分析计算的快速全局优化算法研究[D]. 杭州: 浙江大学, 2007. [12] Sara Carcangiu, Alessandra Fanni. Multi-objective tabu search algorithms for optimal design of electromagnetic devices[J]. IEEE Transactions on Magnetics, 2008, 44(6): 970-973.
2013, Vol. 28 
