Construction of vehicle driving cycle in Fuzhou

Abstract

The current driving cycle in Europe is not consistent with China’s actual road conditions, and the traffic conditions in different cities of China also are discrepant. Therefore, it is increasingly urgent to build corresponding driving cycle based on the driving data of each city. In this paper, based on noise reduction pretreatment of vehicle driving data in three areas of Fuzhou city, K-means clustering is carried out on the first five principal components of the extracted kinematic segments obtained by principal component analysis. Then, three types of kinematic segments, which represent relatively smooth traffic, traffic congestion and smooth traffic, are obtained respectively. So the vehicle driving cycle is constructed, and the correlation coefficient between it and the sampling population for constructing is calculated. Finally, the error analysis method and correlation coefficient are used to verify the constructed driving cycle that is rational and representative.