Gabriel Nicolae CÂRSTOIU, Public Dissertation of PhD Thesis

  Date and Time
Monday, December 13, 2021 - 14:00 to 16:00
  Location
Online

Thesis Title: "Maximizing the management efficiency of rechargeable batteries in critical energy storage applications"

 

PhD Board:

 

  • President: Professor Emeritus Dr. Eng. Vladimir-Ioan CREȚU (Politehnica University of Timisoara)
  • Scientific leader: Professor Habil Dr. Eng. Mihai V. MICEA (Politehnica University of Timisoara)
  • Referees:
    • Professor Dr. Eng. Nicolae ȚĂPUȘ (University Politehnica of Bucharest)
    • Professor Dr. Eng. Sergiu NEDEVSCHI (Technical University of Cluj-Napoca)
    • Professor Dr. Eng. Mircea POPA (Politehnica University of Timisoara)

 

Thesis Summary:

The unprecedented technological progress that mankind enjoys today would not have been possible without certain essential components - rechargeable batteries. Currently, the most used technology is that of rechargeable Lithium-ion batteries. A relatively new class of battery-based applications is that of critical energy storage applications (electric vehicles, backup data storage systems for data centers, etc.). The main topic addressed by this thesis is to reduce the degradation of rechargeable Li-Ion batteries. Although most battery management systems (BMS) in critical energy storage applications implement battery balancing mechanisms, in practice a premature degradation of the component batteries is observed. The reconfigurable connection topology is a recent alternative, but the problem of maximizing the lifespan of these systems, while reducing the total cost of use is incompletely solved. In this thesis we started
from the observation of the analogy between flash memory and reconfigurable battery packs (RBP) to develop a method based on the principles and algorithms of wear leveling, called BWL. The method was integrated into a BMS designed by the author for RBP, and a comprehensive methodology was developed to evaluate the system. By analyzing the BMS performance that integrates the BWL method compared to conventional BMS implementations, its viability and superiority was demonstrated in all scenarios considered. More here.

Note: Persons interested in attending the public hearing may request registration for the online session by e-mail at the address alin.anton (at) cs.upt.ro, at least 72 hours before the date and time set for the support.