CAJ | 학술논문

摘要本文针对混合蛙跳算法（shuffled frog leaping algorithm，SFLA）早熟收敛的问题，将云模型算法融合于SFLA算法中，形成一种云变异蛙跳算法（normal cloud mutation SFLA，NCM—SFLA），弥补混合蛙跳算法后期容易陷入局部最优的不足．同时利用算法易于并行的特点，在多核环境下基于．NET4的并行拓展库（parallel extensions）进行算法的并行优化．将其应用于梯级水库优化调度中，实例计算表明，与多维动态规划算法（MDP）相比，NCM—SFLA方法具有更好的全局寻优能力和较快的收敛速度，在现有的计算条件下该并行算法能有效缩短程序运行时间，求解梯级水库优化调度问题是合理、有效的．
적요본문침대혼합와도산법（shuffled frog leaping algorithm，SFLA）조숙수렴적문제，장운모형산법융합우SFLA산법중，형성일충운변이와도산법（normal cloud mutation SFLA，NCM—SFLA），미보혼합와도산법후기용역함입국부최우적불족．동시이용산법역우병행적특점，재다핵배경하기우．NET4적병행탁전고（parallel extensions）진행산법적병행우화．장기응용우제급수고우화조도중，실례계산표명，여다유동태규화산법（MDP）상비，NCM—SFLA방법구유경호적전국심우능력화교쾌적수렴속도，재현유적계산조건하해병행산법능유효축단정서운행시간，구해제급수고우화조도문제시합리、유효적．
Abstract To improve the premature convergence problem of traditional shumed frog leaping algorithm （SFLA）, in this paper, cloud model algorithm mix together with SFLA algorithm, then a normal cloud mutation shuffled frog leaping algorithm （normal cloud mutation SFLA, NCM-SFLA） is proposed, which is to make up the shortage of shuiffled frog leaping algorithm that is easy to fall into local optimal solution. At the same time, the algorithm is easy to be parallel, parallel extensions are used to parallel optimization of algorithm in multi core environment. Then the algorithms are applied to cascade reservoirs optimal operation. The test case of practical application shows that, compared with the multidimensional dynamic programming algorithm （MDP）, NCM-SFLA has better global search ability and fast convergence speed, and the parallel algorithm can effectively shorten the running time of program in the calculation of existing conditions. Moreover, to solve the cascade reservoirs optimal operation is reasonable, effective by using the new algorithms.