摘要
针对蚁群算法容易陷入局部最优、迭代次数多、稳定性不高的缺陷,提出一种基于禁忌搜索的蚁群改进算法,对初始信息素浓度和信息素更新机制进行优化,并利用禁忌搜索算法的记忆能力和藐视准则,使算法具有跳出局部最优解的能力,同时减少迭代次数。定义"算法相对稳定性"并约定计算规则,用于比较不同算法的稳定性。将改进算法应用于不同城市规模的TSP问题,实验表明,改进算法在寻优能力、迭代次数和稳定性方面的性能均有所提高。
Aiming at the problem that the ant colony algorithm is easy to fall into the local optimum, the number of iterations is high and the stability is fairly low, an ant colony improvement algorithm based on tabu search is thus proposed. The initial pheromone concentration and pheromone update mechanism are optimized. Then the memory ability and contempt criterion of the tabu search algorithm＇s is used to make the algorithm have the ability to jump out of the local optimal solution while reducing the number of iterations. The ＂algorithm relative stability＂ is defined and the calculation rule agreed, and this is used to compare the stability with various different algorithms. The modified algorithm is applied to the TSP problem of different city scales. The experiment indicates that this modified algorithm is clearly improved in terms of optimization ability, iteration number and stability.
出处
《通信技术》
2017年第8期1658-1663,共6页
Communications Technology
基金
贵州省合作计划项目(黔科合计省合[2014]7002])
贵州大学研究生创新基金项目(研理工2016069)~~
关键词
蚁群算法
禁忌搜索
相对稳定性
TSP问题
ant-colony optimization
tabu search
relative stability
TSP problem
