Portfolio item number 1
Short description of portfolio item number 1
Posts by Collection
portfolio
Portfolio item number 2
Short description of portfolio item number 2
projects
Power Electronics Lifetime Estimation
Published:
This project focuses on estimating the remaining lifetime of power electronics modules using machine learning.
publications
Hybrid modeling for remaining useful life prediction in power module prognosis
Published in 2024 25th International Conference on Thermal, Mechanical and Multi-Physics Simulation and Experiments in Microelectronics and Microsystems (EuroSimE), 2024
This paper presents a hybrid approach to estimate the remaining useful life of power electronic modules. It uses Paris law alongside an adaptive polynomial interpolation method to predict the evolution of the module s health indicator
Physics-informed Markov chains for remaining useful life prediction of wire bonds in power electronic modules
Published in Microelectronics Reliability, 2025
In this paper, we utilise kernel density estimation coupled with a Markov chain based sampling scheme to estimate the remaining useful life of power electronic modules. The approach presents high accuracy even when the data is scarce and the prediction task is difficult, namely interpolation and extrapolation.
Data-driven Metamodels for Failure Analysis of Power Electronic Modules
Published in 36th European Symposium on Reliability of Electron Devices, Failure Physics and Analysis (ESREF 2025), 2025
We present the results of using machine learning surrogate models to replace computationally expensive finite element simulations for estimating the remaining useful life of power electronic modules. This approach drastically accelerates the pipeline, achieving a 10⁶ computational speed-up while maintaining high precision with an R² score of 0.962.
AI surrogate models for lifetime prediction in power electronic modules
Published in Microelectronics Reliability, 2026
This work presents an accessible blueprint for applying AI to power electronics, utilizing machine learning metamodels to replace computationally expensive finite element simulations for lifetime prediction. We expand upon this framework by introducing a novel active learning data selection technique to further minimize simulation time, alongside rigorous residual and learning behavior analyses to ensure robust model evaluation.
Data-Efficient Reliability Assessment Using Machine Learning : Application to Lifetime Estimation of Power Electronic Modules
Published in theses.fr, 2026
This thesis introduces data-efficient machine learning methodologies for the lifetime estimation of power electronic modules. Key technical contributions include high-fidelity surrogate models for accelerated damage estimation, a probabilistic Physics-Informed Markov Chain framework, and BEDTime, a few-shot time series forecasting model utilizing Dynamic Time Warping to extrapolate long-term degradation from sparse data.
talks
Hybrid modeling for remaining useful life prediction in power module prognosis
Published:
This work was presented during the 2024 25th International Conference on Thermal, Mechanical and Multi-Physics Simulation and Experiments in Microelectronics and Microsystems (EuroSimE). Its main purpose is to highlight how a physics-based model (Paris’ law) and a data-driven model (adaptive polynomial regression) can be combined to obtain a hybrid model that benefits from both model types’ strenghts.
Estimating the remaining useful life of power electronic modules using aging tests
Published:
This talk was given during the SATIE lab’s annual doctoral students’ day to present my PhD’s work. This talk has recieved the award of the best presentation.
Physics-informed Markov chains for remaining useful life prediction of wire bonds in power electronic modules
Published:
This poster was presented during the 35th European Symposium on Reliability of Electron Devices, Failure Physics and Analysis (ESREF 2024). Throught this poster, we show how kernel density estimation can be utilised as a probability density estimator, to be utilised in a Markov-Chain based sampling scheme. This sampling scheme allows us to capture the dynamic responsible for the power electronic modules’ failure, using experimental data of tests to failure, combined with numerical simulations used to generate mechanical features.
Méthodes data-driven pour l’estimation de la durée de vie restante des modules de puissance
Published:
This presentation was delivered to the STAN team at the LMPS laboratory, bridging the gap between applied machine learning and mechanical engineering. The talk introduced the mechanical community to data-driven approaches for a complex multi-physics problem: estimating the lifetime of power electronic modules. Specifically, it explored how machine learning can be leveraged to model bond wire degradation through a Markov chain-based damage modeling coupled with surrogate machine learning models to infer mechanical indicators based on numerical simulation data.
Combining machine learning with finite element simulations for fast computation in power module failure Analysis due to wire bond degradation
Published:
This talk took place at the 2025 Digital Twins in Engineering & Artificial Intelligence and Computational Methods in Applied Science confernece (DTE - AICOMAS 2025), it presented how machine learning can be used to create surrogate models for finite element simulations to reduce computational time. This work was carreid out to enable advanced frameworks for remaining useful life estimation.
Prognostics of power electronic modules using machine learning
Published:
This poster was presented during the SATIE lab’s annual doctoral students’ day to present my PhD’s work.
L’IA pour la fiabilité des modules de puissance
Published:
Presented to the SATIE team’s branch at Université Gustave Eiffel, this talk was specifically tailored for an audience of power electronics experts. The presentation introduced a machine learning-based perspective on the ongoing challenges of power module reliability. It showcased how data-driven tools, specifically stochastic methods and Dynamic Time Warping algorithms, can be effectively applied to model degradation and improve reliability assessments in this domain.
Data-driven Metamodels for Failure Analysis of Power Electronic Modules
Published:
This presentation explores the use of machine learning-based metamodels to efficiently assess the reliability of power electronic modules, capitalizing on their fast inference speed and high predictive power. A major focus of the presentation is the thorough analysis of often-overlooked statistical properties that are crucial for evaluating model behavior, as well as presenting a novel simulation selection technique based on active learning.
AI 4 Maintenance : Remaining Useful Life Prediction of Power Modules
Published:
This poster was presented during the Mobilit.Ai forum to showcase the adaptation of several machine learning techniques such as active learning and dynamic time warping in the context of lifetime estimation of power electronic modules to develop predictive maintenance strategies.
Data-Efficient Reliability Assessment Using Machine Learning: Application to Lifetime Estimation of Power Electronic Modules
Published:
This presentation represents my official PhD defense, addressing the critical challenge of predicting the lifetime of power electronic modules to enable effective predictive maintenance.
teaching
Mathematical tools for management
Undergraduate course, IUT Sceaux, 2023
This course introduces basic notions of mathematics to undergraduate students (L1 level), specialising in management. As a teaching assistant, my job was to make students familiar with their course materials by going through various exercices and creating an environment that promotes discussions and questions to solidify the students’ knowledge. I was the teaching assistant of 2 groups of 25 students for the first semester and one group of 25 students for the second semester. The topics adressed during this course are mentioned below :
Numerical tools for management
Undergraduate course, IUT Sceaux, 2023
This course highlights how the Central Limit Theorem can be utilised to create confidence intervals and to conduct statistical tests. I was the teaching assistant of a group of 25 undergraduate students (L2 level) during the second semester of the 2023/2024 academic year.