New research on Data-driven Fault Diagnosis for Cyber-Physical Systems

As Industry 4.0 transforms manufacturing through smart, interconnected systems, the complexity of these environments increases vulnerability to faults, posing safety and financial risks.  Mehdi Saman Azari’s thesis addresses these challenges by enhancing fault diagnosis using advanced machine learning, self-adaptive models, and digital twin technologies.

Three main objectives guide the study:

1. developing a robust diagnostic framework resilient to fault pattern variability and data imbalance,
2. enabling knowledge transfer across domains using digital twins and transfer learning, and
3. building a self-adaptive system to handle unseen and evolving working conditions.

The proposed solutions significantly improve fault diagnosis accuracy and adaptability, bridging the gap between research and real-world industrial deployment.

When being asked how he feels about presenting his research at the defense Mehdi Saman Azari says:

- A Ph.D. journey is more than an academic pursuit—it is a transformative experience, marked by challenges, discoveries, and growth. As Omar Khayyam wisely said, “The moving finger writes; and, having writ, moves on.” A Ph.D. is an ongoing process of learning, writing, and progressing. Every challenge faced, every paper written, and every lesson learned contributes to a larger journey that shapes not only the research but also the researcher.

During his PhD-studies Mehdi Saman Azari has been supervised by Francesco Flammini (main supervisor during the beginning of the studies), Jesper Andersson, Linnaeus University (main supervisor at end of the studies) and Mauro Caporuscio, Linnaeus University and Stefania Santini, University of Naples Federico as co-supervisors.

For more information about the defense: https://lnu.se/mot-linneuniversitetet/aktuellt/kalender/2025/disputation-i-data--och-informationsvetenskap-mehdi-saman-azari-20250422/