Le projet ENERGETIC, financé par l’Union Européenne, organise sa première école d’été du 9 au 11 juin 2025 à l’Université de Technologie de Tallinn (TalTech), en Estonie. Cet événement gratuit, en format hybride, s’adresse aux étudiants en master, doctorants, jeunes chercheurs et professionnels.

Sous le thème « IA, batteries et durabilité », l’école propose conférences, ateliers pratiques et échanges autour de l’innovation dans les systèmes intelligents de gestion des batteries, en soutien à la transition verte et numérique de l’Europe.

Parmi les sujets abordés :

• Modélisation et seconde vie des batteries maritimes
• Deep learning pour estimer l’état de santé (SOH) et de charge (SOC) des batteries
• Applications de la spectroscopie d’impédance avec l’IA
• Vision par ordinateur et détection d’anomalies thermiques
• Blockchain et données synthétiques pour les batteries de véhicules électriques
• Réseaux de neurones informés par la physique (PINNs)
• Analyse du cycle de vie (ACV) et durabilité des batteries

Les participants bénéficieront des interventions d’experts académiques et industriels, ainsi que d’exemples concrets et de sessions pratiques.

Inscription gratuite, en ligne ou sur place – inscription sur le code QR ci-dessous ou disponible dans ce flyer.

Training the energy transition: ENERGETIC Summer School to spotlight AI-driven battery innovation in Tallinn

The EU funded project ENERGETIC is pleased to announce its first Summer School, taking place from 9 to 11 June 2025 at Tallinn University of Technology (TalTech) in Estonia. This three-day event is designed to strengthen the skills of early-stage researchers and professionals in the rapidly evolving fields of artificial intelligence, battery technologies, and sustainable energy systems.

With the theme “AI, Battery & Sustainability”, the Summer School offers a deep dive into advanced topics central to ENERGETIC’s mission: accelerating innovation in intelligent battery management systems to support Europe’s green and digital transitions.

Insight, exchange and emerging expertise

The programme, hosted by TalTech, combines lectures from leading academics and industry experts with applied sessions exploring the latest tools and methods for battery health monitoring, data management, and lifecycle engineering.

Participants will explore:

• Modelling and second-life applications for maritime batteries
• Deep learning for estimating battery health and charge
• Impedance spectroscopy and AI across sectors
• Computer vision and anomaly detection in thermal imaging
• Blockchain and synthetic data approaches for EV battery data
• Physics-informed neural networks (PINNs) for smart charging
• Life cycle engineering in the context of sustainability

Building capacity for Europe’s clean energy future

This training initiative is part of ENERGETIC’s broader commitment to high-quality dissemination and skills development across the battery and AI research communities. By equipping the next generation of scientists and engineers with interdisciplinary tools, the Summer School contributes to both capacity-building and the long-term impact of project results.

Open to master’s students, PhD candidates, early-career researchers and professionals, the event is free to attend, with advance registration required.

It is a hybrid event which you can attend onsite or online.

Register using the QR code below or in the following FLYER!

LIST OF SPEAKERS

Assoc. Prof. Henrik Andersen, University of Southern Denmark

Title : Battery modelling & second life in maritime applications – a deep dive

Abstract : In this session, we look into the intricacies of battery modeling and the second life of batteries in maritime applications. The course starts with an exploration of standard battery modeling techniques, focusing on states of a battery, including State of Charge (SOC) and State of Health (SOH). We then transition to Electrochemical Impedance Spectroscopy (EIS) testing and its application in battery modeling, with a specific focus on assessing SOH. The session ends with a case study of an E-ferry’s 4.2 MW battery pack, where EIS modeling is applied to estimate the SOH of batteries in a real-world maritime setting. Through detailed examples and practical applications, participants will gain valuable insights into the challenges and opportunities of applying battery modeling techniques to enhance the performance and longevity of maritime battery systems.

Slimane Arbaoui, PhD student, INSA Strasbourg, France

Title : Use of advanced deep learning models for estimating the State of Health (SOH) and State of Charge (SOC) of lithium batteries

Abstract: The first part of the presentation will introduce the concepts of SOH and SOC, explaining their significance and how they are computed. I will then present techniques for analyzing and preparing data for model training. Following this, I will introduce two deep learning models, LSTM and CNN, demonstrating how they can be used to encode data and make predictions while also providing insights into their working mechanisms.

In the second part, I will address the black-box nature of these models and the importance of explainability. I will introduce Explainable AI (XAI) techniques, specifically SHAP and LIME, and provide a simple algorithm to compute SHAP values. If time permits, I will also discuss counterfactual explanations.

Prof. Yannick Le Moullec, Tallinn University of Technology, Estonia

Title : Overview of Sustainable and Circular Electronics

Abstract : In an age where electronic devices have become ubiquitous, the electronics industry stands at a crucial crossroad, facing the urgent need to address its environmental impact while trying to create enablement effects (sustainable electronics and electronics for sustainability). This lecture will explore sustainable and circular electronics, highlighting key trends, challenges, and opportunities, including joint efforts of industry, academia, regulatory bodies, and other stakeholders. The lecture will provide an overview of such aspects with examples of organizations, bodies, directives, regulations, industry initiatives, and emerging trends in the domain of sustainable electronics. The lecture will conclude with a hands-on session using small footprint resource-constrained edge devices.

Dr. Olev Märtens, Tallinn University of Technology, Estonia

Title : AI and impedance spectroscopy: from healthcare to health of batteries

Abstract : Electrical impedance spectroscopy (EIS) is often an efficient method to characterize the tissues, materials, electrochemical objects and processes and much more, statically or dynamically directly or through inductive, capacitive, electroacoustical or other coupling. Adding machine learning (ML) can significantly enhance the possibilities of this technology.  Use cases so far and challenges are described

Marcella ASTRID, postdoctoral researcher, University of Luxembourg

Title : Anomaly detection in computer vision and thermal imaging for battery monitoring

Abstract : This lecture begins with an introduction to computer vision and its concept and general applications across different fields. It then explores anomaly detection in computer vision, covering its applications, key concepts, and common methods such as one-class classification and pseudo anomaly augmentation. The final section focuses on using anomaly detection in thermal imaging for battery monitoring systems to improve safety and performance.

Dr. Bassem Sellami, Tallinn University of Technology, Estonia

Title : Blockchain and SDN-enhanced architectures for secure and resilient battery data management in EV

Abstract : This session introduces advanced, blockchain-integrated architecture developed within the ENERGETIC project, aiming to enhance security, resilience, and efficiency in battery data management for electric vehicles (EVs). The proposed architecture leverages Blockchain and Software-Defined Networking (SDN) technologies to secure battery management systems (BMS) data transactions, optimize network resources and enable real-time battery monitoring and predictive maintenance. By integrating edge, fog, and cloud computing layers, the architecture ensures robust protection against cyber threats, minimizes latency and improves system scalability and reliability. Empirical results show this approach outperforms traditional blockchain consensus algorithms in transaction throughput, latency, and energy efficiency. Through practical examples, participants will explore the specific advantages and potential challenges of utilizing Blockchain technology in EV infrastructures, thus gaining valuable insights into secure and adaptive battery data management solutions.

Dr. Chahinez Ounoughi, Tallinn University of Technology, Estonia

Title : GAN-powered synthetic data for battery research

Abstract : In this session, we explore the power of Generative Adversarial Networks (GANs) in creating high-quality synthetic data for battery systems. With real-world battery data often being scarce or expensive to collect, GANs provide a powerful solution for generating diverse and realistic datasets. This talk covers the fundamentals of GANs, their application in battery research, and how synthetic data can enhance predictive modeling, anomaly detection, and performance optimization. Through examples and case studies, participants will gain insights into the advantages and challenges of using GANs in battery technology.

Siddhant Dutta, student, University of Mumbai, India

Title : Using Physics-informed neural networks (PINNs) for EV charging power curve aggregation and parameter estimation

Abstract : In this tutorial, we explore the application of Physics-Informed Neural Networks (PINNs) to model and aggregate electric vehicle (EV) battery charging power curves. We demonstrate how to integrate physical principles, such as electrochemical and thermodynamic dynamics, into neural networks to enhance the accuracy and reliability of EV charging predictions. Through hands-on guidance, we show how PINNs can be leveraged to estimate key parameters for optimizing charging infrastructure and improving grid integration. This tutorial serves as a step-by-step guide to understanding and implementing PINNs for EV charging modeling, combining machine learning with physics for smarter energy solutions.

Dr. Bambang Priyono, University of Indonesia

Title : Life cycle engineering of EV batteries

Abstract : The presentation examines the Life Cycle Assessment (LCA) of electric vehicles (EVs), with a focus on lithium-ion batteries. It outlines environmental impacts across the battery life cycle—from raw material extraction to end-of-life—and highlights battery recycling as a key strategy for sustainability. The session emphasizes the relevance of EU regulatory frameworks and explores the case of Indonesia, where lithium scarcity makes battery recycling critically important. Opportunities and challenges are discussed in the context of policy, technology, and stakeholder collaboration. The talk concludes with research directions aimed at advancing circular economy principles and improving the environmental and economic performance of EV battery systems.

Funded by the European Union. Views and opinions expressed are however those of the author(s) only and do not necessarily reflect those of the European Union or CINEA. Neither the European Union nor the granting authority can be held responsible for them. Contract No. 101103667