论文部分内容阅读
雾霾对人类的日常生活带来极大的危害,因而分析产生雾霾的关键影响因素尤为重要.针对目前传统算法预测雾霾关键影响因素存在的缺陷,从一维细胞自动机入手,提出了一种以基于群落弱连接机制的二元萤火虫算法(CWLBGSO)为搜索策略,粗糙集为评价准则的混合方法.CWLBGSO基于自然界中萤火虫间协同进化的“弱连接”机制,划分搜索空间,为每个子空间分配相应的种群,各子种群中的次优个体相互交互产生新个体,从而保持种群的动态多样性,然后将CWLBGSO结合粗糙集,应用于北京,广州和上海三地雾霾关键影响因素的预测中,并结合10交叉验证和SVM算法对预测结果分类准确率和影响因素进行分析,通过与其它算法进行对比,结果表明本文算法能有效剔除冗余因素,预测结果具有较高的稳定性和可行性.
Therefore, it is very important to analyze the key influencing factors of haze.Aiming at the shortcomings of the traditional algorithms in predicting the key influencing factors of haze, this paper proposes a one-dimensional cellular automata A binary firefly algorithm based on community weak connection mechanism (CWLBGSO) is used as the search strategy and the rough set is a mixed method of evaluation criteria.CWLBGSO classifies the search space based on the “weak connection” mechanism of co-evolution between fireflies in nature, The corresponding populations are assigned to each subspace, and the suboptimal individuals in each sub-population interact with each other to generate new individuals to maintain the dynamic diversity of the population. The CWLBGSO combined with rough sets is then applied to the key haze in Beijing, Guangzhou and Shanghai The results show that the proposed algorithm can eliminate redundant factors effectively and the prediction results have a high level of predictive accuracy in the prediction of influencing factors and the combination of 10-cross-validation and SVM algorithm to the classification accuracy and influencing factors of prediction results. Stability and feasibility.