精英交叉二进制蝙蝠算法求解0-1背包问题

发布时间：2018-02-15 22:42

本文关键词： 蝙蝠算法精英策略个体交叉贪心策略柯西变异　出处：《计算机应用研究》2017年06期 　论文类型：期刊论文

【摘要】：针对基本蝙蝠算法收敛速度慢、易早熟的问题,提出了一种精英交叉二进制蝙蝠算法。该算法借鉴精英策略和遗传算法中的交叉机制,按照一定比例选择蝙蝠群中的精英个体进行交叉,将得到的子蝙蝠群和父蝙蝠群进行混合择优,保证蝙蝠群的多样性和优秀性,提高了全局搜索能力;为提高局部搜索能力,算法在对每个个体计算适应度值时加入贪心策略;另外,通过对蝙蝠群最优解进行动态监测,适时对种群进行柯西变异,使算法具有跳出局部极值的能力。通过对五个实例的仿真计算比较表明,该算法与改进贪心遗传算法、贪心二进制蝙蝠算法和病毒协同蝙蝠算法相比,无论是收敛速度还是寻优能力都表现优异,为求解0-1背包问题提供了一个实用的算法。
[Abstract]:Aiming at the problem of slow convergence and premature convergence of the basic bat algorithm, an elite crossover binary bat algorithm is proposed, which draws lessons from the elite strategy and the crossover mechanism of genetic algorithm. In order to improve the local search ability, select the elite individuals in the bat colony to cross according to a certain proportion, and select the sub-bat group and the father bat colony to select the best, to ensure the diversity and excellence of the bat colony, and to improve the global search ability. The algorithm adds greedy strategy to calculate the fitness of each individual. In addition, by dynamically monitoring the optimal solution of the bat population, Cauchy mutation of the population is carried out at the right time. Compared with the improved greedy genetic algorithm, the greedy binary bat algorithm and the virus cooperative bat algorithm, the algorithm is compared with the improved greedy genetic algorithm, the greedy binary bat algorithm and the virus cooperative bat algorithm, and compared with the improved greedy genetic algorithm, the greedy binary bat algorithm and the virus cooperative bat algorithm. Both convergence speed and optimization ability are excellent, which provides a practical algorithm for solving 0-1 knapsack problem.
【作者单位】：河北地质大学信息工程学院;
【基金】：河北省高等学校科学研究计划资助项目(ZD2016005) 河北省自然科学基金资助项目(F2016403055)
【分类号】：TP18

【参考文献】