CAJ | 학술논문

本文针对简单遗传算法的缺陷，设计了一种混合型搜索策略对算法进行改进。这种改进的算法归一化处理了复杂的约束条件，利用精英策略和轮盘赌策略选择最优个体，多点交叉和动态的变异操作使得种群保持多样性。通过改进，使得算法更小几率陷入局部最优，仿真实验表明，这种算法在稳定性、收敛精度上得到了较好的效果。
본문침대간단유전산법적결함，설계료일충혼합형수색책략대산법진행개진。저충개진적산법귀일화처리료복잡적약속조건，이용정영책략화륜반도책략선택최우개체，다점교차화동태적변이조작사득충군보지다양성。통과개진，사득산법경소궤솔함입국부최우，방진실험표명，저충산법재은정성、수렴정도상득도료교호적효과。
In this paper ,aiming at the defects of simple genetic algorithm ,we design a hybrid search strategy to improve the algorithm .Normalization of this improved algorithm to deal with the complex constraints ,using the elite strategy and roulette strategy choice the best individual , multipoint cross and dynamic mutation makes to keep population diversity . Through improvement ,makes the algorithm more small chance to fall into local optimum , the simulation experiments show that this algorithm on the stability and convergence accura‐cy obtained better effect .