摘要
文章以合肥市典型道路为例,选取5条城市道路进行数据采集,采用主成分分析与遗传变异改进的蚁群算法(ant colony optimization,ACO)相结合的方法,构建了合肥市典型行驶工况。在划分了运动学片段的基础上,利用主成分分析法对13个运动学特征参数进行降维处理,以排名前3的主成分为聚类因子,用改进的蚁群算法对运动学片段样本进行分类,通过组合类内运动学片段,完成代表性工况的构建,并对代表性工况进行精度分析。研究结果表明,与K-means聚类法、系统聚类法相比,改进的ACO能够有效提高行驶工况的构建精度。
Taking the measure data of five typical urban roads in Hefei City as an example, typical driv- ing cycle（DC） of Hefei City is constructed with the method of principal component analysis and genet- ic variation improved ant colony optimization（ACO） algorithm. Based on the definition of the kine- mat, ic fragments, the method of principal component analysis is used to reduce the dimension of the 13 kinematic feature parameters, and the first three principal components are classified by the improved ACO algorithm. The representative DC is obtained through the combination of the class of kinematic fragments. The analysis of the precision is also conducted. The results show that compared with K- means clustering algorithm, and system clustering algorithm, the improved ACO algorithm can im- prove the precision of DC construction effectively.
出处
《合肥工业大学学报（自然科学版）》
CAS
北大核心
2017年第10期1297-1302,共6页
Journal of Hefei University of Technology：Natural Science
基金
国家自然科学基金资助项目(71431003)
