Scalable constrained attributes/issues about risk, reliability and optimization in large scale battery pack

Abstract

This study focuses on the optimization of scalable battery pack designs, as a scalable battery pack design can tolerate cell failures, implement a fail-safe mode, rapidly reconfigure itself, and adequately supply the load without interruptions. Hot-swapping, bypass mechanisms, configurability, and redundancy are the key attributes of scalable architecture, which enable the batteries to perform the aforementioned tasks. Scalable designs proposed in the literature still have many limitations, such as suboptimal utilization of the battery packs, low practicality, risk of internal short-circuiting of the batteries, increased weight and size of the battery packs, reliability issues under variable loads, and undesired delays in the reconfiguration of switches. The attributes and issues identified above form the basis of this study. First, the concept of practicality is highlighted for realization of scalable designs in real applications. Scalability associated issues, such as switching risk and sub-optimal sizing are explored. The tree algorithm is applied to develop a scalable design with higher power reliability and faster power redistribution. An SOH balancing method is proposed to optimize the utilization of battery pack. Sample of scalable designs and a physics-based electrochemical battery model are used to implement the proposed algorithms. The results confirm that the proposed SOH balancing can prolong the battery life. Additionally, the tree algorithm based design improves the power reliability, and enables fast redistribution of power.