| Abstract |
The standard driving cycles (DCs) used to evaluate spark-ignition engine-based two-wheelers are inadequate for electric two-wheelers (E2Ws). Also, they fail to accurately represent the actual driving circumstances in specific areas, resulting in inaccuracies during the evaluation of performance. The current research is centred towards constructing an electric two-wheeler urban driving cycle (E2WUDC) that considers the driving circumstances of the smart city in India. Further, the denoised speed data is utilized to extract the micro-trips and compute their driving parameters. Furthermore, the dimensions of the data are decreased through the utilization of principal component analysis. Subsequently, the data is classified utilizing various clustering methods including k-means, X-means, hierarchical clustering, and density-based spatial clustering of applications with noise (DBSCAN). Then, the Calinski Harabasz index (CHI), Davies-Bouldin index (DBI), and silhouette score are used to assess the homogeneity and completeness of selected clustering algorithms in the data cluster. Overall, the E2WUDC is developed using X-means which is selected as a suitable clustering algorithm based on the performance indices. Also, the key driving features of E2WUDC such as total time duration and distance are 14.49 km and 1914 seconds with average and maximum driving speeds of 8 and 13.88 m/s respectively. Eventually, it establishes the foundation for assessing the energy economy, driving range and energy demand for the widespread deployment of electric two-wheelers in urban commuting. © 2024 IEEE. |
