CAJ | 학술논문

针对双链量子遗传算法具有收敛速度慢，容易陷入局部最优解等问题，提出一种新的双链量子遗传算法。该算法将种群个体分组，相邻组间有一个共有个体，即第i组的最后一条染色体同时是第i+1组的第一个个体。组内各染色体同方向同步长更新，相邻组间通过共有个体保持组间同步。该方法能很好地降低算法时间复杂度，保持种群个体的多样性，从而避免算法陷入局部最优值。最后通过实验验证该算法对复杂函数的优化结果明显优于双链量子遗传算法。
침대쌍련양자유전산법구유수렴속도만，용역함입국부최우해등문제，제출일충신적쌍련양자유전산법。해산법장충군개체분조，상린조간유일개공유개체，즉제i조적최후일조염색체동시시제i+1조적제일개개체。조내각염색체동방향동보장경신，상린조간통과공유개체보지조간동보。해방법능흔호지강저산법시간복잡도，보지충군개체적다양성，종이피면산법함입국부최우치。최후통과실험험증해산법대복잡함수적우화결과명현우우쌍련양자유전산법。
Aiming at the problem that the double chains quantum genetic algorithm has low conver-gence rate and is easy to fall into local optimum value, a new quantum genetic algorithm is proposed. In this algorithm, the population is divided into several groups, and there exists a common member between the two neighboring groups. For example, the last member of the ith group is also the first member of the (i+1)th group. All the members of the same group are updated in the same direction with the same step. At the same time, through the common member, the neighboring groups keep pace with one another. This method can reduce the time complexity, maintain the population diversity, and avoid making the algorithm fall into the local optimum value. The simulation results show that the algorithm is much more efficient in the optimization of complex functions than double chains quantum genetic algorithm.