Advanced search

This article tackles the complex challenge of predicting magnetic iron losses in high-frequency magnetic components by introducing a novel conditional generative adversarial network model. Diverging from traditional loss prediction methodologies that often overlook intricate interactions of factors, our conditional generative adversarial network framework is designed to comprehensively incorporate diverse aspects such as material properties, geometrical variations, and environmental conditions. To facilitate this advanced approach, a specialized four-wire measurement kit was employed, which significantly enriched the training dataset with a wide range of measurements. When benchmarked against conventional deep neural network models, the conditional generative adversarial network not only achieves faster convergence but also demonstrates markedly superior accuracy in predicting iron losses. This superiority is particularly notable in scenarios that extend beyond the training data's range, underscoring the model's robustness and adaptability. Such advancements in predictive accuracy and efficiency represent a significant leap forward in the design and optimization of high-frequency magnetic components.

This paper presents a non-invasive, real-time fault detection and predictive maintenance framework for Connected Electric Truck powertrain. Leveraging the ultra-low-power ISM330DHCX sensor’s Machine Learning Core (MLC), the system performs on-sensor processing of accelerometer and gyroscope signals using Edge AI and TinyML techniques. Decision Trees (DTs) and an eight-DT based Random Forest (RF) model were employed to classify vibration patterns under varying operating conditions. Five experiments were conducted on electric motor test benches to simulate rotor imbalance and gather data in both normal and faulty states. Vibration signals were first classified into three broader fault categories and then mapped into 12 subclasses using a novel output mapping scheme. Guided by the Design, Implementation, Potential Failure and Functional Failure (DIPF) curve, this framework enables accurate detection of rotor unbalance faults. Experimental results showed that while single-DT models achieved high training accuracies (95−98.42%) but poor validation performance (<50%), the eight-DT based Random Forest with subclass splitting significantly improved generalization, achieving 98.74% validation accuracy. By minimizing power consumption and bandwidth requirements, this on-sensor Edge AI sensor network approach offers a scalable solution for predictive maintenance, with the potential to reduce downtime and maintenance costs by 30−50%. The findings highlight the promise of ultra-low-power intelligent sensing for enhancing the reliability and efficiency of Connected Electric Trucks.

The Dual Active Bridge Converter topology is widely recognized for its high power density in high-power applications, enabling soft switching and achieving high efficiencies in both buck and boost operation modes. However, under conventional phase-shift modulation, operation at light or no load results in hard-switching and high effective currents, leading to increased overall losses, one of its main drawbacks. These issues have been primarily addressed by implementing complex modulation strategies, leveraging from the multiple degrees of freedom in the control of the converter power, particularly the inner and outer shift angles of its bridges. Contrary to the traditional approach, this work proposes the modulation of the series inductance of the DAB converter by implementing a switched inductor, aiming for a simplified modulation strategy. The proposed method effectively achieves zero current under no-load conditions and significantly reduces effective currents at light loads compared to the traditional phase-shift modulation approach. Although an in-depth comparison with other modulation schemes is required, this work represents a stepping stone in the analysis of the topology and the comprehension of its trade-offs.

This article explores battery health monitoring in electric vehicles (EVs) using machine learning to address challenges in battery durability and enable new business models. It introduces a virtual battery prototype that applies supervised learning methods, such as Random Forest and Deep Neural Network regression, to estimate real-time energy slack and monitor battery health. The study also presents a carbon balance optimization application, aiming to minimize carbon emissions and charging costs for EV fleets through grid optimization. The model enables continuous battery health monitoring, opening opportunities for innovative commercial use cases for EV users, fleet managers, and grid operators.

This report identifies the critical research and innovation (R&I) priorities for decarbonising Europe's heavy-duty vehicles, based on direct feedback from industry stakeholders. The findings reveal a consensus: battery electric technology is the primary pathway forward, with significant stakeholder support for R&I focused on its improvement. While battery electric technology is perceived as more mature, hydrogen is considered a complementary solution for the most demanding long-haul routes. Large-scale demonstrations are suggested for de-risking operations and evaluating integration with the transport and energy system. The analysis confirms that achieving TCO parity or better compared to diesel is the most important factor for market uptake. This study provides direct, evidence-based guidance for EU transport R&I policy, helping to chart the road ahead and orient R&I call programming to meet the ambitious CO₂ emission standards for heavy-duty vehicles.

This report presents a comprehensive overview of the activities carried out in Work Package 4 (WP4), specifically Task 4.3, of the RHODAS project, which focuses on the design and development of a digital twin framework for key components of electric powertrains.  The developed framework comprises three individual digital twins: the inverter digital twin, the e-motor digital twin, and the gearbox digital twin. The inverter digital twin models the electrothermal behavior of power electronics modules (SiC and GaN) to support performance monitoring and failure prediction. The e-motor digital twin captures the thermal behavior of the rotor and stator of the RHODAS e-motor, enabling accurate simulation under various driving conditions. The gearbox digital twin is based on a data-driven thermal equivalent circuit model that estimates the gearbox oil temperature using a range of input data.  All digital twin models are implemented in both MATLAB/Simulink and C++ to enable seamless integration into the RHODAS cloud platform, supporting both online and offline thermal monitoring and efficient computation.  The deliverable also includes extensive sensitivity analyses to investigate the behavior of the e-motor, power inverter, and gearbox under different operating modes. This outcome delivers a robust and scalable digital twin architecture that enhances monitoring, diagnostics, and predictive maintenance capabilities within the RHODAS electric vehicle platform. 

DENSO Corporation, based in Japan, is a global leader in automotive technology and components. The company specializes in the development and production of advanced systems and products for vehicles, including powertrain control systems, thermal systems, electronics, and advanced safety technologies. Founded in 1949, DENSO operates in over 200 locations worldwide and serves major automakers, including Toyota, Honda, and General Motors.

This deliverable presents the design, validation, and testing of the RHODaS high‑power hybrid T‑type multilevel inverter. Chapter 2 explains the overall architecture of the inverter, including sensors, modular power stages, mechanical structure, housing, and integration. The first design of the power stage did not meet the electrical and thermal requirements; therefore, several improvements were introduced in the final stage, which are also described in this chapter. Both iterations, the initial and the final design, are documented to highlight the evolution of the system. Chapter 3 explains the initial T‑type design and the challenges encountered with the first GaN transistors, such as voltage limitations, short‑circuit behaviour, and reliability issues. Chapter 4 explains the initial inverter tests, including switching behaviour and operation at different load points, which were performed to ensure proper functionality. Finally, chapter 5 defines the comprehensive high‑power inverter tests, covering efficiency, thermal performance, and maximum power capability, with final validation to be conducted at BOSMAL’s mechanical testing laboratory. In conclusion, the deliverable documents the progression from an initial design with critical shortcomings to a robust final inverter prototype, achieving a power density of 58.6 kW/l. The initial tests confirm that the converter is both reliable and capable of operating in line with the project specifications, reaching efficiencies of up to 99%. Nevertheless, the definitive validation of the converter will be conducted at BOSMAL, where the comprehensive test procedures defined in this deliverable will be applied to ensure full compliance with the project requirements. 

In this study, the basic structure of the pure electric command vehicle is studied, the main components of the command vehicle power system, namely the selection of the drive motor and the power battery, are analyzed, and the main parameters of the drive motor and the power battery are designed and calculated. The calculation results show that the power and transmission system developed in this paper meets the design requirements, and the design scheme is feasible and reasonable.