Abstract
Li-ion batteries have gained increased popularity in the past few decades as the main source in various mobile and stationary energy storage applications. Battery management system design, especially fault diagnosis, however, is still a challenge regarding Li-ion batteries. Traditional Li-ion BMSs rely on measurements from current, voltage, and temperature sensors sparsely located throughout the battery pack. Such a BMS is not capable of predicting battery behavior under various operating conditions; moreover, it cannot account for internal discrepancies among battery cells, incipient faults, the distributed nature of battery parameters and states, and the propagation effects inside a battery pack. Although majority of these effects have already been observed and reported, they are either studied in electrochemistry laboratories using in-situ techniques and detailed theoretical analysis or in practical manufacturing settings by engineers and technicians, which are typically considered proprietary information. The aim of this paper is to bridge the gap between these two domains. In other words, a detailed electrochemical/thermal simulation of a Li-ion battery cell under healthy and faulty conditions is performed to provide a better understanding of the exact spatial requirements for an efficient and reliable thermal management system for Li-ion batteries. The results of this study are specifically of great importance for battery fault detection and identification, mainly due to the recent advancements in distributed sensing technologies such as fiber optics.
Calibration Optimization Methodology for Lithium-Ion Battery Pack Model for Electric Vehicles in Mining Applications