CAJ | 학술논문

为了解决基本人工蜂群算法（ABC）早熟收敛、容易陷入局部最优、收敛精度不高等问题，提出一种混合改进的人工蜂群算法（RABC）。首先，为了平衡ABC的全局寻优能力，在初始化种群阶段引入了混沌算子和逆向学习算子；而后，为了提高局部寻优能力，在采蜜蜂的检索方程中引入了最优引导个体；最后，为了提高收敛精度和加快后期收敛速度，改进了侦察蜂的检索机制。为了验证RABC算法的收敛效果，通过在3个标准测试函数上的仿真实验，并与基本ABC算法的比较，发现RABC的收敛性能有显著提高。
위료해결기본인공봉군산법（ABC）조숙수렴、용역함입국부최우、수렴정도불고등문제，제출일충혼합개진적인공봉군산법（RABC）。수선，위료평형ABC적전국심우능력，재초시화충군계단인입료혼돈산자화역향학습산자；이후，위료제고국부심우능력，재채밀봉적검색방정중인입료최우인도개체；최후，위료제고수렴정도화가쾌후기수렴속도，개진료정찰봉적검색궤제。위료험증RABC산법적수렴효과，통과재3개표준측시함수상적방진실험，병여기본ABC산법적비교，발현RABC적수렴성능유현저제고。
In order to solve the problem of the basic Artificial Bee Colony (ABC)algorithm,such as the premature convergence,falling into local optimum easily,low convergence precision,put forward an Revised Artificial Bee Colony (RABC)algorithm.First,in order to balance the ABC global optimization ability,in the initialized population stage introduce the chaos operator and reverse learning operator. Then in order to improve the local optimization ability,in mining bee search equation introduce the best guide in the individual.Finally, in order to improve the convergence precision and speed up the convergence speed,improve the search mechanism of scout bees.In order to verify the convergence effect of RABC,through the simulation experiments on three standard test functions,and compared with the bas-ic ABC algorithm,found that the convergence of the RABC have improved significantly.