Majdi Mansouri is a Tunisian academic and Associate Professor in the Department of Electrical & Computer Engineering at Sultan Qaboos University, Oman. He holds an HDR in Computer Engineering from the University of Orléans, a PhD in Electrical Engineering from the University of Technology of Troyes, and a Master’s degree in Signal and Image Processing from ENSEIRB Bordeaux. He is internationally recognized for his work in artificial intelligence, machine learning, statistical signal processing, and intelligent fault detection systems, with applications in energy systems, industrial monitoring, and smart grids. His research also spans deep learning, Bayesian methods, digital twins, IoT, and computer vision. He has held academic and research positions in France, Qatar, Saudi Arabia, and Oman, including roles at Texas A&M University at Qatar and Prince Sultan University. He is actively involved in teaching courses such as machine learning, signals and systems, and digital logic design, and has supervised numerous PhD and master’s students across multiple countries. He has an extensive publication and editorial record, serving as Associate Editor for IEEE Access and reviewing for several IEEE journals. He has led and contributed to multiple funded research projects, including major grants from Qatar National Research Fund and Sultan Qaboos University, focusing on AI-driven energy systems, fault diagnosis, and predictive analytics. He has been recognized among the Top 2% Scientists worldwide by Stanford University–Elsevier rankings and has received multiple research excellence awards. He is also active in international collaborations, conference organization, and academic leadership in intelligent systems and industrial AI.

Hybrid AI for Fault Diagnosis: Integrating Physics-Based Models, Data-Driven Intelligence, and Multimodel Fusion and Applications to Renewable Energy Systems Modern renewable-energy systems such as wind turbines and solar photovoltaic plants operate under highly variable conditions and exhibit complex, nonlinear failure mechanisms. This talk presents a Hybrid AI framework that unifies physics-based models with advanced data-driven intelligence to deliver reliable, interpretable, and high-accuracy fault diagnosis. Physics-based modeling preserves the fundamental electrical, mechanical, and thermal dynamics of the system, ensuring that diagnostic predictions remain physically consistent. In parallel, data-driven models learn subtle degradation signatures, rare fault behaviors, and operational uncertainties directly from real-world measurements. These two perspectives are integrated through multimodel fusion, enabling the framework to combine complementary hypotheses, reconcile conflicting information sources, and maintain robustness under noisy, incomplete, or heterogeneous data conditions. Applications to wind and solar energy illustrate the practical impact of this Hybrid AI approach. Across both domains, the multimodel Hybrid AI strategy significantly increases fault-detection sensitivity, reduces false alarms, and supports predictive maintenance, ultimately contributing to higher reliability, lower operational costs, and improved energy availability in modern renewable-energy infrastructure .