一种基于模糊关系的联盟演化策略

A Strategy of Coalition Evolvement Based on Fuzzy Relation

提出一种新的联盟演化策略,基于模糊集合论,给出了当前任务和历史任务"等价性"的概念,在不能实现任务完全匹配时,由模糊关系的合成,将问题限制在较小范围内求解,使联盟能快速及时形成.最后,通过试验仿真说明了此策略的有效性.和前人的工作相比,该方法能显著降低联盟形成过程的复杂度,而且演化策略也较为简单易行.

In this paper, a novel strategy of coalition evolvement is presented. Based on fuzzy set theory, definition of ＂equivalence＂ among current task and historic tasks is given, when tasks can not reach a perfect match, the problem is limited in a small scope to be solved and the coalition can be formed quickly and timely by means of composition of fuzzy relations. The effectiveness and advantages of this method is demonstrated by simulation. Compared with previous work, this strategy can drastically reduce the complexity of coalition formation, and is easier to implement.