Battery sizing and performance evaluation of battery electric vehicles

The adoption of electric vehicles (EVs) is a promising solution to reduce air pollution. Batteries serve as the primary energy storage system (ESS) in EVs, and their sizing significantly affects their overall performance. A straightforward method to determine the battery energy size is to integrate the power requirements over a given drive cycle. However, a key challenge arises because of the interdependence between battery energy size and battery mass, which requires an optimization approach. To address this issue, this study proposes an optimized battery sizing method using particle swarm optimization (PSO), selected for its simplicity and computational efficiency. A key novelty lies in the incorporation of traction motor dynamics via a transfer function model, which enables realistic performance estimation during optimization. The proposed method is validated against a conventional iterative sizing approach using both the worldwide harmonized light vehicle test cycle (WLTC) and a real driving cycle (RDC). The results show that PSO reduces the battery size by 4.67%, battery mass by 4.76%, and total cost by 4.75%. While the iterative method achieves a higher driving performance owing to a larger battery, the PSO-sized battery still achieves 224.5 km of autonomy and 10.37 years of estimated battery life under WLTC conditions. In RDC testing, the PSO method yields 270.2 km of autonomy and a 22.38-year battery life. These results demonstrate that PSO enables lightweight and cost-effective battery sizing while maintaining acceptable EVs performance. Future work will focus on multi-objective optimization to better balance costs, energy, and endurance.