HAL Portal UTC Université de Technologie de Compiègne
Not a member yet
11652 research outputs found
Sort by
Impact of Model Depth on State of Charge Estimation in Second-Life Batteries
No full textInternational audienceThis paper investigates the impact of different levels of modelling on the accuracy of state of charge estimation in battery storage systems. The aim is to compare simple and complex models to identify the optimal balance between accuracy and computational complexity. The results show that while deeper models improve estimation, the additional complexity provides only marginal benefits. A kinetic approach improves accuracy at high loads without significantly increasing the effort. Separating the inverter and battery models increases flexibility by allowing system changes without retooling. The results provide a basis for integrating temperature effects and optimising storage operation for economic efficiency
Control of Battery-Integrated MMCs with NLM using Distributed Control Architecture
No full textInternational audienceBattery-integrated modular multilevel converters (BI-MMCs) have several series-connected submodules, each with a small battery module. This design has high cell-level control, potentially improving efficiency and battery lifespan compared to traditional EV battery inverter systems. A distributed control architecture is preferred for BI-MMCs with a large number of submodules, and in such control architectures, the communication network limits the sample rate of the reference signal. For variable speed drives, where the fundamental frequency can be up to a few kHz, the reference sampling frequency can adversely affect the submodule battery current quality and accuracy. The concept of reconstructing the reference signal at a higher sampling frequency than the communication link's update rate for BI-MMCs using nearest level modulation is introduced and validated through simulations and experimental results
Control of a Modular Multiport Solid State Transformer for a Flexible High Power Charging Infrastructure
No full textInternational audienceThis paper presents a control strategy for a Multiport Solid-State Transformer designed to connect multiple EV charging stations to the medium-voltage AC grid. The transformer employs a Cascaded H-Bridge converter composed of identical switching cells, each incorporating a Dual Active Bridge to provide galvanic isolation between the AC grid and the charging ports. Steady-state modelling of the multilevel converter is used to identify the permissible power imbalance between charging ports, ensuring stable operation. The proposed real-time control method facilitates independent output power control for each port while maintaining effective cell voltage balancing across the converter. Furthermore, dynamic switch matrix reconfiguration enhance the converter's operational range. The control strategy is validated through simulations and experimental measurements, demonstrating improved flexibility and performance for scalable EV charging infrastructure
Low inductance power module optimized for Flying Capacitor Topology
No full textInternational audienceIn Renewable Energy domain, power converter require adapted solutions to increase electrothermal performances. In this paper, a high-performance packaging architecture is proposed, dedicated to a flying capacitor topology. Based on 750V SiC bare dies, interconnection technologies, materials and optimized design was established to obtain a dedicated power module, with the support of finite element modeling. A specific assembly process was developed with innovative tooling and was tested at the end of process by electrical static tests and Xray analysis. Power electrical performances was measured by double pulse test method. Electrical characterization and module simulation recovered the value of inductance close to 6nH. Next step is to evaluate the robustness of this new power module with campaigns of passive temperature cycling and power cycling to be performed
Containerized AC/DC Converter Station as a Building Block for Medium-Voltage DC Grids
No full textInternational audienceDC grids are one solution to help balance the growing energy demand with distributed renew- able generation and storage systems. To simplify the setup of medium-voltage (MV) DC grids, a simple system architecture with few standardized core components is necessary. For the coupling of new MVDC grids to existing AC grids, this paper proposes a design for a modular multilevel converter (MMC) station that could be preman- ufactured in factory and easily shipped to any site as a building block with up to 40-MW power rating. To achieve the required level of integra- tion and compactness, innovative approaches for component and system-level design need to go hand in hand, as will be illustrated in the paper
Operation modes and design consideration for MMC with unipolar current full-bridge submodules
No full textInternational audienceThe unipolar current full-bridge submodule (SM) for modular multilevel converters (MMC) operates with unipolar current and bipolar voltage. Two basic operating modes for unipolar arm currents are presented, described, and analyzed. The differences between the two idealized operating modes are further discussed regarding the system design of an MMC. The design factors for current and voltage in the DClink differ, which leads, for example, to different numbers of SMs and power losses. These aspects are presented as examples and trade-offs are summarized
New DC-Link bus bar and capacitors integration for 800V inverter
No full textInternational audienceThis project was funded by the French State as part of France 2030, and by European Union as part of France Relance. Authors strong gratitude to the R&D and engineering teams of Punch Powertrain, EFI and all partners of Mautiv'8 project, for their invaluable contributions to this development of safe and reliable DC-Link for 800V inverter. Their hard work and expertise have been instrumental in advancing the field
Advanced DC Voltage Control Technique for Converters: Achieving Superior Speed and Stability with Model Predictive Control
No full textInternational audienceAmong different converter topologies, theactive front-end (AFE) rectifier has been widely usedin recent research for generating controlled output DCpower. Simultaneously, model predictive control (MPC)has emerged as a prominent and actively researchedtopic, particularly for its applications in developmentof advanced control strategies. In this paper, a controlmethod is presented that leverages the predictive capabilityof MPC to achieve the desired reference power throughthe DC power generated by the AFE rectifier. A newexponential function has been introduced in the proposedcontrol method as a dynamic reference power for theMPC. This reference power can be updated accordingto the DC-link voltage requirements for an effective DCvoltage regulation. Finally, the performance of the proposedcontrol method is compared with the performanceof the conventional PI-based control method and precedingMPC-based control techniques, to validate its superiorityover the existing control methods. Implemented and verifiedin Matlab Simulink, this technique shows five-timefaster recovery and zero steady state error
Energy transition: an approach from scientific features to societal issues
No full textInternational audienceThis article presents the experience conducted by researchers at the University of Lille on the technical, scientific and societal issues of the energy transition. Thanks to an academic chair on Energy Transition and an «open laboratory», cross-disciplinary views from the academic world, industry, civil society, local authorities and associations were raised and encouraged. The objective is to identify methods for implementing the energy transition that take into account citizens' aspirations, environmental constraints and technological limitation altogether
Failure Causality Diagnostic in Industrial Systems through Automated Machine Learning
No full textInternational audienceIn the context of modern industrial systems, efficient failure management is crucial for maintaining operational integrity, minimizing downtime, and maintenance optimization. This paper explores the application of Automated Machine Learning (AutoML) to enhance both failure causality diagnostic and failure causality prognostic in industrial systems. Different failure causes are detected by failure causality diagnostics, and the upcoming failure could be prevented by failure causality prognostics. Indeed, upcoming failures could be avoided by preventing their causalities.Traditional machine learning approaches require significant manual intervention for model selection, hyperparameter tuning, and feature engineering, which can be time and cost-consuming. AutoML, on the other hand, automates these processes, enabling quicker and more accurate predictions while reducing the need for extensive domain expertise.Our approach integrates AutoML into real-time failure diagnostics, identifying the root causes of system malfunctions using historical and sensor data. Simultaneously, it applies AutoML for prognostics, predicting Remaining Useful Life (RUL) of components and foreseeing future failures. By leveraging both data-driven models and physics-based insights, our approach improves the reliability of diagnostics and prognostics in various industries, including manufacturing, aerospace, and energy.The competing risks can be considered an application of this approach. High probable competing risks to cause are detected based on historical data in diagnostic phase and by handling them in the prognostic phase, the anticipated failures could be prevented.Finally, the experiments are conducted using real-world industrial datasets, demonstrating the superior performance of AutoML compared to traditional machine learning methods in both diagnostic accuracy and prognostic precision. This study shows that AutoML can significantly enhance decision-making processes in maintenance planning and risk mitigation, ultimately reducing operational costs and improving system reliability