Skip to Main Content
The majority of design and controller optimization methodologies for electric vehicles (EVs) are based on single-cycle optimization strategies. It has been shown that design parameters that are optimized for one average-assumed drive cycle (DC) are not necessarily optimal when applied to different DCs or to the entire driving profile of a vehicle. Stochastic optimization techniques that consider a multitude of DCs in their objective function can overcome the suboptimalities associated with this so-called cycle beating but, as a drawback, require a large amount of stochastically representative DC data. This paper presents a novel methodology to generate stochastic DCs for the design and control optimization of EVs. DCs and driving profiles are segmented into modules of similar modes of vehicle operation. The key physical parameters of each module are identified and stochastically analyzed. The obtained probability functions of each key parameter are then implemented in a DC generation tool (DCGT). The DCGT is capable of generating an unlimited amount of DCs by reassembling the DC modules according to their stochastic composition. The results show that DCGT-generated DCs accurately represent the original DC data with respect to the frequency spectra, speed distribution, acceleration distribution, and load characteristics of their correspondent duty cycles.