Multi-objective optimal design of lithium-ion battery packs based on evolutionary algorithms

• An optimization methodology for the battery thermal management design is proposed.
• The methodology is based on multi-objective PSO and multi-physics simulations.
• A theoretical case shows the trade-off between temperature operation and area.
• A real battery pack based on pouch cells for a solar car was designed.
• A novel battery packaging design framework is able to find better solutions.

Lithium-battery energy storage systems (LiBESS) are increasingly being used on electric mobility and stationary applications. Despite its increasing use and improvements of the technology there are still challenges associated with cost reduction, increasing lifetime and capacity, and higher safety. A correct battery thermal management system (BTMS) design is critical to achieve these goals. In this paper, a general framework for obtaining optimal BTMS designs is proposed. Due to the trade-off between the BTMS's design goals and the complex modeling of thermal response inside the battery pack, this paper proposes to solve this problem using a novel Multi-Objective Particle Swarm Optimization (MOPSO) approach. A theoretical case of a module with 6 cells and a real case of a pack used in a Solar Race Car are presented. The results show the capabilities of the proposal methodology, in which improved designs for battery packs are obtained.