(518c) Uncertainty Analysis in Lithium Ion Battery Packs

Lithium ion batteries have been used in increasingly high power applications over the past decade. With an operating window of 4 V and providing among the highest of energy densities available today, lithium batteries are a competitive choice for a variety of applications ranging from laptop computers to electric vehicles. Even with such attractive capabilities, the demand for higher energy and power batteries has lead to larger cell sizes and assembly of larger size battery packs. When the energy requirements of the system are cannot be met by a single cell, battery makers typically resort to assembly of several cells in a series-parallel configuration. Examples of such large size packs are the batteries used in satellites that need to meet precise energy requirements and the automotive batteries that have high power requirements as well. In all these applications, long battery life is a typical requirement, and extensive efforts towards experimental testing of cell performance as well as theoretical prediction of battery life have been undertaken. Such efforts in the past, however, did not effectively accommodate for the differences among the individual cells that constitute the pack. Consequently, such results from controlled tests cannot reliably predict the performance of a battery pack employed in a practical application. In this work we propose to incorporate robustness analysis as part of the mathematical models used predicting battery life.