EU-funded projects like ESCALATE often provide updates about deliverables, events or other contributions. Thus, the researchers and their work, which happens “behind the scenes”, are often not showcased. That’s why we decide to put our PhD students from RWTH Aachen into the spotlight by regularly sharing interviews about their research and the connection with ESCALATE. These conversations are published at regular intervals, which will start with several questions addressed to Christoph Wellmann, who works on Mechatronics in Mobile Propulsion.
Please describe your research to a wider audience? On what kind of innovative aspects are you currently working on?
Our research involves developing standardised, modular, and scalable electric powertrain systems, specifically targeted at heavy-duty long-haul trucks. Central to our innovation is an advanced e-axle architecture, combining optimised electric motors and gearboxes into one compact package. We apply an automated co-optimisation approach that simultaneously adjusts the electrical design of permanent magnet synchronous motors (PMSMs), transmission gear ratios, and control strategies. This approach uses active learning algorithms and numerical optimisation, significantly enhancing performance and efficiency. For instance, in our tests, we achieved a peak torque of 56,150 Nm, surpassing the required target by about 2%. We also incorporated innovative cooling strategies and lightweight materials like optimised aluminium housing, significantly reducing energy losses and vehicle weight. These solutions provide major improvements in energy efficiency, durability, and driving comfort compared to traditional systems.
What are the connections with the ESCALATE project? How do you benefit from the project, and what synergies exist?
Our research is part of the ESCALATE project, which allows us to access to comprehensive benchmarking studies covering major truck manufacturers and powertrain suppliers, which identified critical trends and needs, guiding our modular e-axle development. Furthermore, ESCALATE provides valuable industry-standard targets and real-world constraints, ensuring our designs are practical, robust, and cost-effective. In return, our cutting-edge optimisation methodology contributes directly to the ESCALATE objectives, helping standardise and scale electric powertrain technology across different vehicle classes. This mutual support accelerates innovation, promotes component standardisation, and ultimately reduces costs for future commercial applications.
With your expertise and the perspective of your research, where do you currently see the biggest challenge in a large-scale deployment of zero-emission trucks?
Based on our findings, one major challenge for large-scale deployment of zero-emission trucks is managing the integration of high-performance electric powertrain components within existing vehicle architectures. Our research highlights that fitting powerful electric motors and gearboxes within tight packaging constraints poses significant engineering challenges, especially given stringent durability requirements like achieving bearing lifetimes above 10,000 hours. Another notable challenge involves ensuring reliability under real-world driving scenarios; our simulations revealed that achieving optimal energy efficiency (around 109.8 kWh per 100 km for 40-ton trucks and 225 kWh per 100 km for 76-ton trucks) requires sophisticated e-axle design and predictive cruise control strategies.
What do you think could be the impact (or lack thereof) of projects like ESCALATE on the promotion of zHGV technology/solutions?
Projects such as ESCALATE have considerable potential to positively influence the development and adoption of zero-emission heavy-duty vehicle (zHGV) technology. By fostering collaboration among a wide array of stakeholders, including leading OEMs, technology providers, and researchers, ESCALATE directly accelerates innovation and standardisation. Its approach of developing modular, scalable, and standardised powertrain components significantly reduces development time and cost, enhancing industry readiness. Importantly, our research shows that such projects offer clear performance benchmarks (like the achieved peak wheel torque of 55,000 Nm and optimal energy usage in standardised cycles), essential to convincing stakeholders about the viability of zero-emission solutions.
How can the described machine learning-integrated methodology support the development process, and what specific benefits in terms of frontloading capabilities and modularity arise from this?
The described machine learning-integrated methodology greatly enhances the development process by combining automated optimisation with active learning techniques. This integration allows engineers to rapidly predict and refine powertrain component designs, significantly accelerating iterative cycles that previously depended on manual tuning. Specifically, the approach enables frontloading capabilities, meaning critical design decisions – such as electric motor sizing, bearing selection, and transmission ratios – are validated early through simulations rather than costly late-stage prototypes.
This reduces development risks, costs, and time-to-market by ensuring technical feasibility at initial stages. Additionally, this methodology inherently supports modularity, as trained ML models can adapt quickly to varying design requirements across different vehicle classes. For example, parameters optimised for one variant are efficiently leveraged to predict performance in others, greatly simplifying customisation and component reuse. Ultimately, this ML-driven strategy promotes standardised, flexible, and scalable solutions, aligning closely with industry needs for rapid, cost-effective, and adaptable powertrain development.