摘要
利用Logistic微分方程建立了GA种群规模的数学模型;为更好地使初始解群覆盖整个解空间,引入Logistic周期-混沌迭代公式;针对GA在求解优化问题时,种群早熟问题,将生物学中培育野生个体和避免近亲交叉的概念引入GA中,进一步提高GA中个体基因的多样性。为提高GA在迭代后期收敛速度,利用Tabu搜索的爬山能力加快GA收敛。计算机仿真算法验证所提算法的有效性,并与标准GA进行比较。
A model of race colony for GA is presented based on Logistic differential equation. The known Logistic iterative-equationof period-chaos is introduced into GA to cover whole solution space by initial solution group. The eoncept of wildlife that is hybridizedwith tamed life to cultivate new breed is developed from biology to GA to restrain the early maturing of iterative process in GA. The conver-gence of iterative process during late period is accelerated by Tabu Search (TS). Computer simulation result on an example of the optimiza-tion of unit operation in hydropower station has shown that the Logistic GA (LGA) based method is more efficient than the standard GA(SGA) based method.
出处
《中国电力》
CSCD
北大核心
2004年第9期42-45,共4页
Electric Power
基金
天津市重点自然科学基金(023801211)
