Liens vers les séminaires de 2025, 2024, 2023, 2022, 2021, 2020, 2019, 2018, 2017, 2016, 2015, 2014, 2013, 2012, 2011, 2010
Caiyi Xiong
(LIAS, Poitiers)
04/06/2026 12h30-14h en salle Chaussey au H9 (ECL) : Real time data driven model learning of complex dynamical systems application to high-speed rotating machines operated with active magnetic bearing
Active Magnetic Bearings (AMB) are widely used in high-speed rotating machines due to their ability to levitate rotors without mechanical contact. The considered AMB-supported rotor system is a multivariable system that is open-loop unstable, which makes system identification particularly challenging and requires identification to be performed in closed loop. This article aims at developing data-driven identification methods to track the evolving dynamics of AMB systems in real time, with the objective of detecting performance degradation and anticipating potential failures.
At the current stage of the research, the system is assumed to be time-invariant and the objective is to obtain an initial characterization of its dynamic behavior, in particular the resonance and anti-resonance information. For this purpose, a nonparametric identification is carried out in the frequency domain. Multisine excitation signals are applied to the system and only steady-state responses are retained. Several estimators were evaluated, and the Errors-In-Variables (EIV) estimator was selected to estimate the frequency response Gk. This step provides an accurate nonparametric representation of the system dynamics. Based on this result, future work can be extended to the identification of parametric models.
Francisca Viejo Reales
(University of Huelva, Spain)
03/06/2026 12h30-14h en salle Chaussey au H9 (ECL) : Parameter Identification of a Dynamic Nitrification Model Using Experimental Soil Solution Data
Human activities, especially the widespread use of synthetic fertilizers, have profoundly disrupted the natural nitrogen cycle. This disruption has led to severe environmental issues, including eutrophication, biodiversity loss, and the formation of marine dead zones. According to the Planetary Boundaries Framework, the safe threshold for nitrogen input into the environment has already been largely exceeded. Addressing this imbalance is, therefore, a necessity. In this context, optimizing fertilization is a key step toward restoring balance and moving towards more sustainable agricultural systems.
To achieve this goal, it is essential to gain a deeper understanding of the behavior of nitrogen species in soil. One of the key processes in this cycle is nitrification, a biological process in which ammonium is converted into nitrite and then into nitrate through the action of microorganisms (bacteria). In this regard, the development of models of nitrification represents a valuable tool to describe and predict the dynamics of the process. Nevertheless, the parameters of such models are highly sensitive to multiple environmental and biological factors, reflecting the inherent complexity of living systems. This preliminary work aims at identifying a set of parameters of a dynamical model of the nitrification process using a experimental dataset consisting in concentration data of nitrogen species obtained under controlled laboratory conditions.
Maté Kiss
(TU Eindhoven)
20/05/2026 12h30-14h en salle Chaussey au H9 (ECL) : Data-Driven Model Completion : from experimental design to interpretable models
The quality of a model resulting from (black-box) system identification is highly dependent on the quality of the data that is used during the identification procedure. Designing experiments for linear time-invariant systems is well understood and mainly focus on the power spectrum of the input signal. Performing experiment design for nonlinear system identification on the other hand remains an open challenge as informativity of the data depends both on the frequency-domain content and on the time-domain evolution of the input signal. Furthermore, as nonlinear system identification is much more sensitive to modelling and extrapolation errors, having experiments that explore the considered operation range of interest is of high importance. Hence, this work focuses on designing space-filling experiments i.e., experiments that cover the full operation range of interest, for nonlinear dynamical systems.
Abdelghani GHANAM
, chercheur contractuel CNRS au sein de l’équipe IS2E
30/04/2026 12h45-13h45 à lLieu : ECL, Amphi 202 – Bâtiment W1 : Graphène Induit par Laser : De la synthèse à l’intégration d’électrodes dans les systèmes électrochimiques
Graphène Induit par Laser : De la synthèse à l’intégration d’électrodes dans les systèmes électrochimiques
Résumé : Le graphène induit par laser (Laser-Induced Graphene, LIG) est une structure carbonée 3D poreuse et conductrice, obtenue en exposant un précurseur riche en carbone à un faisceau laser, sans réactifs chimiques ni étapes de purification. En jouant sur les paramètres laser, la composition du précurseur et l’atmosphère de travail, il est possible de contrôler finement la morphologie et les propriétés électrochimiques du matériau obtenu, voire de synthétiser in situ des matériaux hybrides (oxydes métalliques, matériaux 2D). Cette présentation illustrera comment le LIG peut être intégré comme électrode fonctionnelle dans des systèmes variés : capteurs électrochimiques, stockage d’énergie, électrocatalyse, valorisation de la biomasse et piles à combustible microbiennes. Le LIG se positionne ainsi comme une plateforme simple et versatile pour le développement d’interfaces électrochimiques hautes performances.
Mots-clés : Graphène induit par laser, électrodes fonctionnelles, électrocatalyse, ·stockage d’énergie, systèmes électrochimiques.
English version :
Laser-Induced Graphene : From Synthesis to Electrode Integration in Electrochemical Systems
Abstract : Laser-Induced Graphene (LIG) is a porous, conductive 3D carbon structure obtained by exposing a carbon-rich precursor to a laser beam, without chemical reagents or purification steps. By tuning the laser parameters, precursor composition, and working atmosphere, it is possible to finely control the morphology and electrochemical properties of the resulting material, and even to synthesize hybrid materials in situ (metal oxides, 2D materials). This presentation will illustrate how LIG can be integrated as a functional electrode in a variety of systems : electrochemical sensors, energy storage, electrocatalysis, biomass valorization, and microbial fuel cells. LIG thus stands as a simple and versatile platform for the development of high-performance electrochemical interfaces.
Keywords : Laser-induced graphene, functional electrodes, electrocatalysis, energy storage, electrochemical systems.
Maxime Grosso
(en post-doc actuellement au sein de l’équipe CoSMa)
23/04/2026 12h45-13h45 à l’INSA de Lyon, bâtiment St Exupéry, 1e étage, salle de réunion : Quand les harmoniques deviennent des états : modélisation et commande des systèmes périodiques, application à la machine synchrone à réluctance variable
Les systèmes à coefficients périodiques sont omniprésents en génie électrique : machines tournantes, convertisseurs, réseaux AC, centrales inertielles... mais ils peuvent mettre en défaut les approches de la commande linéaire classique, qui négligent les couplages inter-harmoniques et les effets périodiques. Le formalisme du Harmonic State Space (HSS) (ou des Dynamic Phasors - DP) offre un cadre rigoureux pour résoudre ce problème : il transforme exactement un système temps périodique (TP) en un modèle temps invariant (TI) de dimension infinie, dont les états sont les harmoniques prises individuellement des signaux. Pour les systèmes LTP, il permet la mobilisation directe des outils de la commande LTI classique (retour d’état, LQR, H∞, LMI, LPV...) dans le domaine harmonique, et d’en déduire une commande elle-même à coefficients périodiques avec des garanties de stabilité. Ce séminaire détaillera la théorie HSS et illustrera la démarche par la synthèse d’un retour d’état périodique visant l’élimination des harmoniques de couple d’une machine à réluctance variable (SynRel).
Mots clés :
Harmonic State Space · Dynamic Phasors · Linear Time-Periodic systems · LMI · Robust control · Torque ripple · Harmonic Mitigation
Références clés :
G. Floquet, "Sur les équations différentielles linéaires à coefficients périodiques," Annales scientifiques de l’École Normale Supérieure, 1883.
S. R. Sanders et al., "Generalized Averaging Method for Power Conversion Circuits," IEEE Transactions on Power Electronics, 1991.
E. Mollerstedt and B. Bernhardsson, "Out of Control Because of Harmonics : An Analysis of the Harmonic Response of an Inverter Locomotive," IEEE Control Systems, 2000.
N. Blin et al., "Necessary and Sufficient Conditions for Harmonic Control in Continuous Time," IEEE Transactions on Automatic Control, 2022.
- P. Riedinger and J. Daafouz, "Solving Infinite-Dimensional Harmonic Lyapunov and Riccati Equations," IEEE Transactions on Automatic Control, 2023.
F. Vernerey, P. Riedinger, and J. Daafouz, "A TBLMI Framework for Harmonic Robust Control," IEEE Transactions on Automatic Control, 2025.
Contributions :
M. Grosso, "Modélisation & Commande Harmonique Temps Réel pour les Chaînes d’Actionnement Électrique," Thèse de Doctorat, 2025. tel-05505787
M. Grosso et al., "Harmonic Control of Three-Phase AC/DC Converter With Time-Domain Guarantees," IEEE Transactions on Control Systems Technology, 2025. hal-04161410
M. Grosso et al., "Control of three-phase PWM rectifier using multiple frame dq transform and harmonic modeling," IECON, 2024. hal-04866357
M. Grosso et al., "Frequency-Varying Harmonic Domain Control for PMSMs with Current Harmonic Mitigation," ECCE, 2025. hal-05327724
M. Grosso et al., "Harmonic Modeling and Control under Variable-Frequency," Preprint, 2026. hal-05535041
M. Grosso, P. Riedinger, and J. Daafouz, "The PhasorArray toolbox," ECC, 2026. hal-05327912
Nihal KULARATNA
21/04/2026 13-14h à l’INSA de Lyon, bâtiment Léonard de Vinci, 3e étage : Can supercapacitor assisted converter and protection techniques change the road map of future DC systems ?
Séminaire organisé dans le cadre de l’Ecole Doctorale EEA de Lyon.
With the world moving towards renewable energy resources AC versus DC debate comes back. Rapidly developing power electronic systems help us exchange power across DC and AC interfaces. State of the art power converters used to link DC output based renewable systems to grid come with several issues : (i) complexity of the product (ii) limited conversion efficiency (iii) requirement of central energy storge systems (ESS). When these converters keep feeding excess energy back to the grid it brings the potential issue of grid-stability.
Given that solar/wind outputs are DC, industry has chosen to combine SPWM based inverters with MPPT based DC-DC converters and battery banks. Today, lighting, whiteware and infotainment industries develop consumer products where line frequency disappears in the rectifier blocks, where AC line frequency or voltage regulation is not a concern. This case justifies moving into a DC world dominated by DC (micro) grids and DC powered consumer products. In DC systems, rectifier losses are primarily eliminated. Most ‘inverter driven” white goods and lighting systems do not require a precise DC rail. By combining these two advantages, together with battery-free distributed energy storage, electrical industry can justify the advantage of DC systems, with (a) increased efficiency (b) less components for reliability (c) less battery storage for environmental friendliness.
After 2000, supercapacitor (SC) industry has grown with over 1200 manufacturers and over 60k employees and a growth rate of 2.83%. Today single cell SCs come in four families with device sizes from fractional farads to over 100,000 F. Given that their equivalent series resistances are one order lower than lithium based rechargeable batteries SCs allow very high charge/discharge currents without overheating.
These high temperature capable SCs combined with ohms order loop resistances they create large time constant circuits. For example, 1 F SC combined with a 1-ohm loop resistance creates a circuit with a time constant (τ) of 1 second. This 3V rated SC comes with a storge capacity of 4.5 joules, compared with a 1 µF, 100V capacitor with a storage capacity of only 5 mJ forming a τ of 1µs. A common 3 V rated SC of 3000F comes with an energy storge capability of 13.5 kJ or 3.75 Wh. A larger SC of 100,000 F could store over 65 Wh. This practical scenario leads to consider these devices to be used in power converters with built in short/medium term energy storage and transient over voltage and over-current protection systems. These new Supercapacitor Assisted (SCA) converters and protection systems combined with DC powered consumer systems help us eliminate the three main issues (i) to (iii).
Presentation is on the new Supercapacitor Assisted (SCA) design approach for more efficient, less complex and central energy storge free future converter systems and protection techniques for the DC world. SCA techniques presented will include EMC free low speed DC-DC converters, DC LED lighting, DC whiteware, surge protectors and DC circuit breakers, with a brief introduction to an experimental household hybrid inverter system based on distributed SC energy storage.
Plus d’informations :
