论文部分内容阅读
采用大规模种群进化优化策略，根据用户评价时间与单一数值适应值估计用户评价个体模糊适应值；根据个体表现型属性与参照个体模糊适应值宽度计算个体表现型相似度；利用个体表现型相似度对种群聚类并估计未评价个体模糊适应值；基于个体模糊适应值和表现型相似性构造个体选择适应值，实现个体相似性选择。将所提方法应用于室内挂钟进化设计, 并与已有典型方法比较。结果表明, 所提方法在提高优化质量、减轻用户疲劳、提高搜索效率和实际应用等方面均具有优越性。
The large-scale population evolutionary optimization strategy was used to estimate the individual fuzzy fitness value of the user based on the user evaluation time and the single fitness value. The individual similarity was calculated according to the individual phenotype attribute and the reference individual fuzzy fitness value. Clustering the population and estimating the fuzzy fitness of the unapproved individuals. Based on the individual fuzzy fitness and phenotypic similarity, we construct the fitness of individual selection to achieve individual similarity selection. The proposed method is applied to the design of indoor wall clock evolution, and compared with the existing typical methods. The results show that the proposed method is superior in improving the quality of optimization, reducing user fatigue, improving search efficiency and practical application.