摘要
为提高遗传算法的寻优能力,引入模拟自然界蜂王繁殖的改进型遗传算法(QEBGA)。概述了QEBGA的实现过程,指出基本遗传算法(SGA)采用轮盘赌选择机制选择种群个体,以普通变异算子对种群作变异操作;而QEBGA采用启发式选择机制选择种群个体,按比例分别以普通变异算子和强变异算子对种群作变异操作。详述了电力系统经济调度问题表述为极小化下的总费用函数及约束优化问题。最后,用6台发电机系统和13台发电机系统的模拟实验比较了QEBGA和SGA 两种算法在优化性能上的差异,实验结果说明了相同种群规模下,QEBCA的寻优时间小于SGA;系统规模越大,QEBCA在计算精度上的优势就越突出。
A Queen- bee evolution based genetic algorithm ( QEBGA ) , which simulates the reproduction of the queen-bee, is introduced to improve the optimization capability of the genetic algorithm. SGA selects individuals by the roulette mechanism and uses common mutation operator, while QEBGA adopts the heuristic mechanism and different mutation operators according to set proportions. The minimal total cost function and constraint optimization for power system economic dispatch are expounded. The simulative experiments for 6-unit system and 13-unit system are carried out to compare the optimization capability between QEBGA and SGA. Results show that, in same population size, the optimal searching time of QEBGA is less than that of SGA, and the larger the system is, the better the precision of QEBGA shows.
出处
《电力自动化设备》
EI
CSCD
北大核心
2005年第5期64-66,共3页
Electric Power Automation Equipment
基金
国家自然科学基金资助项目(NSFC 50007002)
