I am a doctoral researcher jointly affiliated with IFP Energies Nouvelles (IFPEN) and CRAN (CNRS, Université de Lorraine). My research addresses a practical gap in process monitoring: how to detect and diagnose equipment faults in chemical pilot plants where operating conditions shift deliberately and frequently — an environment where conventional fault detection methods struggle to distinguish genuine degradation from intentional process variation.
My work spans reconstruction-based anomaly detection, temporal deep learning, and graph neural networks, benchmarked on synthetic data and validated on real industrial pilot plant data. A central finding so far is that current data-driven methods function as extreme event detectors rather than sensitive fault monitors under variable operating regimes — an insight driving my development of operating-condition-aware detection frameworks.
Before the PhD, I completed a Master's in Mechatronics, Machine Vision, and Artificial Intelligence at Université Paris-Saclay, with a mobility semester at Poznań University of Technology focused on autonomous systems. My technical range extends beyond process monitoring into computer vision, 3D reconstruction, and generative models. I hold a Bachelor's in Electrical and Electronic Engineering from IGEE (ex-INELEC) in Algeria.
