AAU logo

PhD defence by Vaclav Knap on Characterization, Modelling and State Estimation of Lithium-Sulfur Batteries


06.12.2017 kl. 08.30 - 12.30


Vaclav Knap, Department of Energy Technology, will defend the thesis "Characterization, Modelling and State Estimation of Lithium-Sulfur Batteries".


Characterization, Modelling and State Estimation of Lithium-Sulfur Batteries


Vaclav Knap


Professor Remus Teodorescu


Associate Professor Tamas Kerekes 


Professor Josep Guerrero, Dept. of Energy Technology, Aalborg University (Chairman)
Søren Højgaard Jensen, Department of Energy Conversion and Storage, Risø campus, DTU, Denmark
David A. Howey, University Of Oxford, UK


Lithium-Sulfur (Li-S) batteries represent an appealing battery technology, which might become an alternative for the currently wide spread Lithium-ion batteries. However, the current limitations concerning the cell performance and lifetime are the major factors, which slow down their commercialization. Vast research efforts are carried out to improve the cell design and composition; nevertheless, only a minimum of work has been focused on characterizing their behavior and developing tools for a prospective use in practical applications.

Thus, this thesis has tried to fill in the aforementioned gap and it studied several aspects for understanding the Li-S batteries behavior for prospective practical applications. Therefore, the equivalent electrical circuit model for discharging of the Li-S batteries has been developed. The proposed model is able to simulate the dynamic voltage response of the studied Li-S battery cell during the discharge and is also suitable to be used for extended Kalman filter based state-of-charge (SOC) estimation. Moreover, an equivalent electrical circuit approach was also used for the analysis of the electrochemical impedance spectroscopy performed at the Li-S cell under various temperature and state-of-charge levels.

An extensive experimental laboratory testing procedure was used to characterize the Li-S cell and to develop its models. The short-term self-discharge was experimentally investigated in detail and a simple, but effective model was proposed for it. Based on this self-discharge model, the self-balancing feature of the Li-S batteries was identified, which enhances their balancing and might even lead to avoid an additional electronic circuitry usually implemented for this purpose. Furthermore, the charge recovery effect and thermal attributes were investigated. Because of the specific behavior of Li-S batteries, a specially tailored testing methodology to evaluate the battery’s performance parameters change during the ageing was proposed. The testing methodology covers the characteristic behavior such as the cumulative history, rapid self-discharge and it also includes the measurements of the unique polysulfide shuttle current.

The work was also focused on state estimation, which is an important functionality of battery management systems. Recursive Bayesian filters: an extended Kalman filter, an unscented Kalman filter and a particle filter, were used for the SOC estimation of the Li-S cell based upon the voltage response of the cell and the model. The SOC estimation approach was further improved by implementing the dual extended Kalman filter, where the first filter identified online the battery model parameters, and the second filter based on these parameters estimated the SOC. This approach was also successfully accommodated for the battery state-of-health estimation in terms of capacity fade and change of the internal resistance. Furthermore, the online identified parameters were used in the dynamic model based approach to estimate the maximum available power of the battery during the dynamic use.


PhD defence by Vaclav Knap on Characterization, Modelling and State Estimation of Lithium-Sulfur Batteries





Department of Energy Technology


Pontoppidanstræde 111, auditorium