摘要
遗传算法存在着早熟的现象,易陷入局部最小点,为了克服这个缺点,笔者提出改进云自适应遗传算法(improved cloud adaptive genetic algorithm,ICAGA),即将云模型引入遗传算法,由X条件发生器自适应调整交叉变异概率,使交叉变异概率既具有传统自适应遗传算法的趋势性,满足快速寻优,又具有随机性,改善避免陷入局部最优能力,然后用模拟农夫捕鱼算法(SFOA)中的收缩搜索来对云自适应遗传算法进行修正,以获全局最优解。以网损最小为目标函数,对标准IEEE 14和IEEE 30节点系统进行仿真计算,结果表明该算法能够获得更好的优化解。
Genetic algorithm (GA) has the defect of prematurity during the optimization process which makes it easy to fall into the local minimum. To cope with this defect, the author proposes improved cloud adaptive genetic al-gorithm (ICAGA). Bying applying GA in the cloud model, the adaptive adjustment oi me crossover and mutation probability can be done by X condition generator, which enjoys both traditional AGA trend and the fast optimization with randomness. The algorithm can aslo improve the capacity to avoid falling into local optimal. Then the idea of shrinking search in the Simulating Fisher fishing Optimization Algorithm (SFOA) is brouhgt in to obtain the global optimaization. Taking the minimum network loss as objective function, the simulation for the proposed ICAGA algo-rithm by standard IEEE 14 -bus system and IEEE 30 -bus system are performed. The simulation results show that the better optimal solution can be attained by the proposed ICAGA algorithm.
出处
《黑龙江电力》
CAS
2015年第4期347-352,共6页
Heilongjiang Electric Power
关键词
云理论
网损最小
云自适梯度遗传算法
无功优化
cloud theory
minimum network loss
roposed ICAGA algorithm. cloud adaptive genetic algorithm
reactive power optimization