As part of the TEAMING project, I spent four months at Politecnico di Torino working on adaptive control strategies for EV traction inverters. My research aims to improve energy efficiency by analyzing experimental data under various operating conditions and switching frequencies.

The electrification of transport is reshaping the future of mobility. As electric vehicles (EVs) become more widespread, improving the efficiency and reliability of traction systems is essential for achieving longer driving ranges and lower energy consumption. However, EV traction systems operate under highly variable speed and load conditions, which makes maintaining high efficiency across all operating points a significant challenge.
My research focuses on developing adaptive control strategies for inverter operation optimization. Instead of relying on fixed parameters or switching frequencies, the control algorithm aims to dynamically adjust itself based on real-time operating conditions. The goal is to minimize total system losses while keeping the torque ripple within acceptable limits. Such adaptive control concepts can help make EV powertrains more sustainable and energy-efficient without requiring hardware modifications.
To strengthen the experimental foundation of this work, I carried out a four-month secondment at Politecnico di Torino (PoliTo) under the TEAMING project. The stay was hosted by Professor Radu Bojoi’s Lab, a state-of-the-art facility for advanced electric drive research. The main objective of the secondment was to move from simulation-based analysis to experimental data collection and validation—investigating how inverter control strategies behave under realistic operating conditions on a full-scale EV traction platform.
During this period, we conducted a series of experiments using a PI+SVPWM control strategy across multiple switching frequencies and DC-link voltages. The tests provided valuable insights into how switching behavior influences inverter and motor losses, overall system efficiency, and torque quality. These results are currently being analyzed to identify efficiency trends and trade-offs under different speed/torque/Vdc combinations, forming the foundation for further optimization studies.
Beyond data collection, this secondment also provided essential experience in real-time control implementation using platforms such as dSPACE and Typhoon HIL. Collaborating closely with PoliTo researchers helped bridge the gap between theoretical models and experimental practice, and inspired new directions for future work. One promising direction is the use of AI-assisted optimization, which could automatically identify the most energy efficient switching frequency for each operating condition based on the experimental database. This approach may pave the way for next generation adaptive control frameworks that combine physical modeling with intelligent decision-making.
I am deeply grateful to AAU Energy and Politecnico di Torino for their continuous support and guidance, and to the TEAMING project for enabling such meaningful cross-institutional collaboration. The knowledge and experience gained from this stay have laid a strong foundation for the next steps of my research on intelligent inverter control and its contribution to the future of smart, energy-optimized electric vehicles.

Ying He

Aalborg University – Politecnico di Torino