Sara Sattarzadeh
Published: 2022
Total Pages: 0
Get eBook
Among the available energy storage devices, Lithium-ion Batteries (LIBs) are showing significant promise for many applications such as renewable power grids, Electrified Vehicles (EVs), and consumer electronics due to their high energy density, long life, and lack of memory effect. However, safety is still one of the critical barriers to lithium-ion battery technologies. From a battery control viewpoint, real-time diagnostics of battery faults is a key towards safer batteries. These battery faults can originate from many factors, such as manufacturing defects, abusive operating conditions, and internal degradation mechanisms induced by aging. Therefore, early detection of such faults at their nascent stage is indispensable for battery safety. This dissertation proposes fault diagnostics techniques based on system theoretic approaches to improve the safety of batteries by considering various aspects of safety. In the first sub-problem, we present a computationally efficient battery model that captures individual electrode-level behavior in LIBs. Such electrode-level control can effectively expand the battery cells' usable energy and power limits by utilizing the knowledge of individual electrodes' charge and health. Furthermore, internal degradation mechanisms can be identified utilizing such electrode-level information. Second, irrespective of the physical cause of the failure, many internal faults eventually manifest themselves as abnormal thermal behavior, which may, in turn, lead to thermal runaway. Therefore, in this dissertation, the thermal safety in Lithium-ion batteries is aided by a combination of installed temperature sensors and thermal management algorithms. We propose a framework that finds sensors' effective locations that maximize state observability and proposes a real-time algorithm for distributed temperature estimation in pouch cells. In third sub-problem, we propose a framework that (i) optimizes the sensor locations to improve the detectability and isolability of thermal faults in pouch cells, and (ii) designed a filtering scheme for fault detection and localization based on a two-dimensional thermal model. In the last sub-problem, we propose a closed-loop feedback based approach that enables real-time optimal charging protocol adaptation to battery health, and posses active diagnostic capabilities in the sense that it detects real-time faults during charging and takes corrective action to mitigate such fault effects.