Component Sizing Optimization Based on Technological Assumptions for Medium-Duty Electric Vehicles

In response to the stipulations of the Energy Policy and Conservation Act and the global momentum toward carbon mitigation, there has been a pronounced tightening of fuel economy standards for manufacturers. This stricter regulation is coupled with an accelerated transition to electric vehicles, catalyzed by advances in electrification technology and a decline in battery cost. Improvements in the fuel economy of medium- and heavy-duty vehicles through electrification are particularly noteworthy. Estimating the magnitude of fuel economy improvements that result from technological advances in these vehicles is key to effective policymaking. In this research, we generated vehicle models based on assumptions regarding advanced transportation component technologies and powertrains to estimate potential vehicle-level fuel savings. We also developed a systematic approach to evaluating a vehicle’s fuel economy by calibrating the size of the components to satisfy performance requirements. We used Autonomie, a high-fidelity vehicle modeling and simulation tool developed by Argonne National Laboratory, integrating Pattern Search solvers to optimize component sizing based on our assumptions. Pattern Search, a direct-method numerical optimization algorithm, is widely used in a variety of applications. The method requires extensive evaluation and iteration but provides good optimization performance for the computational cost. This paper presents the potential energy savings for a medium-duty electric vehicle determined using both rule-based and optimized component sizing.