基于偏好排序淘汰NSGAII算法的短波网络多区域重点覆盖优化方法

发布时间：2019-05-24 09:15

【摘要】：在采用偏好NSGAII算法求解多子区域重点覆盖的短波网络频率优化指配时,针对算法中非支配排序耗时较多的问题,该文提出一种偏好排序淘汰的NSGAII算法。在进行非支配排序前,根据解的偏好评价排序结果淘汰一部分偏好评价较差的解,减少参与非支配排序的解的数量从而减少求解时间,同时降低偏好评价结果较差的个体解被选中进行交叉、变异的概率,提高算法的求解效率和求解效果。在进行的48组数据测试中,该文算法在其中38组决策解偏好评价结果和求解时间同时最优,相同迭代次数时相比偏好NSGAII算法节省27%的求解时间。结果表明通过偏好排序淘汰机制的引入,更好利用了偏好信息,使算法用较少的时间求得更好的偏好解。
[Abstract]:When the preference NSGAII algorithm is used to solve the frequency optimization assignment of shortwave networks with multiple subregions, a preference sorting elimination NSGAII algorithm is proposed to solve the problem that non-dominant sorting takes a lot of time in the algorithm. Before the non-dominant ranking is carried out, according to the preference evaluation result of the solution, a part of the solution with poor preference evaluation is eliminated, and the number of solutions participating in the non-dominant ranking is reduced, thus reducing the solution time. At the same time, the individual solution with poor preference evaluation results is selected to cross, the probability of mutation is reduced, and the solving efficiency and effect of the algorithm are improved. In the 48 sets of data tests, the preference evaluation results and solution time of 38 groups of decision solutions are optimal at the same time, and the solution time is saved by 27% compared with the preference NSGAII algorithm when the number of iterations is the same. The results show that the preference information is better utilized by the introduction of preference sorting elimination mechanism, and the better preference solution is obtained in less time.
【作者单位】：华南理工大学电子与信息学院;
【基金】：国家自然科学基金(61571192) 广东省公益研究(2015A 010103003)~~
【分类号】：TN925

【参考文献】