摘要
提出一种新的可用输电能力的计算方法.将蚁群优化算法的正反馈特性与实数遗传算法的进化策略相结合,克服了基本蚁群算法只适用于离散问题的局限性,并提高了寻优的效率、全局的寻优能力和结果的稳定性.在计算过程中,根据不等式约束越界量的大小,动态调整罚函数,采用强制搜索策略,提高了算法的收敛速度,有效克服了在计算可用输电能力过程中,可能出现因早熟而陷入局部最优解的问题.以IEEE-30节点系统为例进行可用输电能力的仿真计算,并与其他算法进行比较,结果证明了该算法的合理性、有效性和优越性.
A new computation method of available transfer capability （ATC） was presented. This algorithm combines the positive feedback of ant colony optimization （ACO） with the evolutionary strategy of float genetic algorithm （GA）, and introduces the pattern search method as eugenic strategy, thus enhances the optimization efficiency, global convergence performance and stability of result. Meanwhile, according to the amount overrunning the limit of inequality constraints during the computing process, this algorithm uses the non-stationary multi-stage assignment penalty function to simplify the inequality constrains, and adopts the forced searching strategy to further increase the convergence speed and improve the global optimization results. Compared with the other algorithms, the verification results by IEEE 30-bus system showed the rationality, availability, and superiority of this algorithm.
出处
《浙江大学学报（工学版）》
EI
CAS
CSCD
北大核心
2009年第11期2073-2078,共6页
Journal of Zhejiang University：Engineering Science
基金
国家自然科学基金资助项目(50177004
50977009)
关键词
可用输电能力
蚁群算法
遗传算法
罚函数
强制搜索
available transfer capability
ant colony optimization
genetic algorithm
penalty function
forced searching
