Ph. D. Project
Observer design for multi-cell lithium-ion batteries
2022/10/01 - 2025/09/30
Other supervisor(s):
Prof. Raël Stéphane (
Electrochemical batteries are ubiquitous in our daily lives, whether in our computers or our cell phones.
Among the different technologies available, lithium-ion batteries offer many advantages in terms of energy
mass, power mass and low self-discharge. In addition, they do not have a memory effect. On the other hand,
they require a management system (BMS) for safety reasons, but also to prevent premature aging.

The BMS plays a key role on the battery performance and lifespan. It requires for this purpose to have access
to accurate data on the current state of the battery. The problem is that little information about the internal
variables is directly accessible through measurements, typically the current, the voltage and possibly the
temperature. To access the battery states (state of charge, state of health, functioning state), a mathematical
model of the battery dynamics is usually developed, based on which an observer is designed to estimate the
non-measurable internal variables. Different approaches have been developed for this purpose [1], in
particular by CRAN and GREEN, whose main originality is to exploit local electrochemical models and then to
build non-linear observers with convergence and robustness guarantees [2-4].

However, it appears that most of these approaches are dedicated to the state estimation of a single lithium-ion
cell. However, in practice, batteries are most often composed of multiple cells associated in series and/or in
parallel. For implementation reasons, single-cell approaches cannot simply be duplicated when many cells are
interconnected. It is therefore necessary to develop estimation tools, that are easy to implement, adapted to
multi-cell batteries and which generate guaranteed data.

The objective of this PhD thesis is to develop methodological tools with low computational requirements for
the state estimation of multi-cell lithium-ion batteries. The results obtained will be validated in Matlab
Simulink interfaced with dSPACE using experimental data.

[1] Y. Wang, J. Tian, Z. Sun, L. Wang, R. Xu, M. Li, Z. Chen. (2020). A comprehensive review of battery modeling
and state estimation approaches for advanced battery management systems. Renewable and Sustainable
Energy Reviews, 131, 110015.
[2] P.G. Blondel, R. Postoyan, S. Raël, S. Benjamin, P. Desprez. (2018). Nonlinear circle-criterion observer design
for an electrochemical battery model. IEEE Transactions on Control Systems Technology, 27(2), 889-897.
[3] P.G. Blondel, R. Postoyan, S. Raël, S. Benjamin, P. Desprez. (2017). Observer design for an electrochemical
model of lithium ion batteries based on a polytopic approach. IFAC-PapersOnLine, 50(1), 8127-8132.
[4] E. Planté, R. Postoyan, S. Raël, Y. Jebroun, S. Benjamin, D. Monier Reyes, Multiple active material Lithium-
ion batteries: finite-dimensional modeling and constrained state estimation, 2021.
Control engineering, observer, estimation, Lyapunov stability, batteries, electrochemical modeling,
Candidates must have a M.Sc. in control engineering/theory, applied mathematics or electrical engineering.
Expertise in Matlab is expected.

Duration: 3 years

Starting data: fall 2022
Control Identification Diagnosis
Université de Lorraine, LUE « accompagnement de la dynamique interdisciplinaire »