IPM Machine Design Using K-Means Data Clustering Technique for Automotive Applications

The purpose of this paper is to investigate the challenges that arise while doing an analysis of electric machines throughout the course of a complete driving cycle. Because of the complicated nature of this process, which is caused by the large number of operational points, the use of specialist equipment and the expenditure of a considerable amount of time are both requirements. This study's overarching goal is to improve the efficacy of electric machine design by investigating clustering strategie. The purpose of this study is to investigate various methods of clustering in order to create an electric machine design that is more effective. The proposed approach employs clusters of operating points to identify chosen Representative points or RPs to construct electric machines with the highest possible efficiency within a given operating range. The study suggests the automated k-Means method for cluster analysis and RPs detection. To test the effectiveness of the proposed method, the study conducted an electromagnetic design study and analysis of the internal permanent magnet machine (IPM) for the WLTP driving cycle. As a result, the operating point set was reduced to only eight points, allowing for an assessment of the k-Means technique's efficiency. This study has implications for researchers and practitioners seeking to improve electric machine design and efficiency in the automotive industry.