The transition toward greener and more sustainable energy systems requires efficient, reliable, and robust electromechanical systems. Electrical machines, power electronic converters, and associated energy conversion systems play a major role in transportation, industry, and renewable energy applications. Their reliability is therefore essential to improve energy efficiency, reduce maintenance costs, and extend system lifetime.

This keynote will present recent research activities in the field of condition monitoring, fault diagnosis, and predictive maintenance of electromechanical systems, with a particular focus on green energy conversion applications. The work is based on the analysis of electrical machines, power converters, and complete electromechanical systems under different operating conditions. A multiphysics approach is considered by combining electrical, mechanical information in order to better understand system behaviour and identify fault signatures.

Particular attention is given to the diagnosis of electrical machines and power electronic systems, including multiphase machines, which offer interesting advantages in terms of fault tolerance, reliability, and continuity of operation. These characteristics are especially important for transportation electrification and other critical energy conversion applications.

Advanced signal processing and data analysis methods are used to improve fault detection and diagnosis. Artificial intelligence and machine-learning techniques can also provide useful complementary tools for analysing complex data and supporting fault classification. However, these approaches must remain closely connected to the physical understanding of the system in order to obtain reliable and interpretable results. The research activities carried out within the DIAMME and DIAMME-LAB projects aim to develop and validate advanced diagnostic methods through experimental investigations under realistic operating and fault conditions. The objective is to improve system monitoring, anticipate failures, and support more efficient maintenance strategies.

The keynote will also discuss some important challenges, including variable operating conditions, limited fault data, sensor selection, robustness of diagnostic indicators, and the integration of different physical measurements.

The final objective is to contribute to the development of more reliable, efficient, and sustainable electromechanical systems, supporting the evolution toward greener energy conversion.