论文部分内容阅读
本文研究了粒子群优化的模糊聚类方法在车辆行驶工况中的应用。采用主成分分析方法将众多反映车辆行驶工况特征的运动学片段特征值进行压缩,用粒子群优化的模糊聚类方法对运动学片段的前三个主成分得分进行聚类,通过Matlab编程将上述理论用于合肥市典型道路行驶工况的构建和分析,按时间比例选取合适片段拟合代表性工况,并将代表性工况和采用K均值聚类法及模糊C均值聚类方法拟合的工况进行对比分析。研究结果表明,将粒子群优化的模糊聚类方法应用到工况的构建中可以有效地提高构建精度。
This paper studies the application of particle swarm optimization fuzzy clustering method in vehicle driving conditions. Principal component analysis (PCA) is used to compress a large number of eigenvalues of kinematic segments that reflect the characteristics of vehicle driving conditions. The first three principal component scores of kinematic segments are clustered by particle swarm optimization fuzzy clustering. The above theory is used to construct and analyze the driving conditions of typical roads in Hefei. The appropriate sections are selected according to the time proportion to fit the representative working conditions. Representative working conditions and K-means clustering and fuzzy C-means clustering Combined conditions for comparative analysis. The results show that the application of particle swarm optimization fuzzy clustering method to the construction of conditions can effectively improve the construction accuracy.