Seminar by Nassima Aissani

An integrated AI-IoT framework for safety and reliability in the context of the energy transition: research and prospects

The transition to Industry 4.0 represents a profound paradigm shift that is radically redefining industrial safety, equipment maintenance and environmental management, particularly in the emerging and strategic field of green hydrogen. 

Our research focuses on an integrated framework for anomaly detection and condition monitoring, tailored to a prototype hydrogen production system utilising both solar and wind energy. 

This framework synthesises research covering system modelling, safety analysis, machine learning-based equipment condition monitoring and real-time data acquisition via the Internet of Things (IoT). 

The convergence of key Industry 4.0 technologies — IoT sensors, modern communication protocols (MQTT, Node-RED), artificial intelligence, cloud computing and augmented reality — and advanced methodologies in reliability engineering and risk analysis is explored through practical case studies.

This integrated approach offers a powerful, data-driven methodology aimed at improving the safety, reliability and operational efficiency of renewable hydrogen production by combining rigorous safety modelling with advanced machine learning techniques, thereby supporting predictive maintenance strategies that reduce unplanned downtime, lower maintenance costs and mitigate safety risks. 

These findings ultimately facilitate the deployment of resilient, efficient and safe hydrogen production systems capable of supporting a sustainable energy future.

Professor Nassima Aissani is a lecturer and researcher at the Institute of Maintenance and Industrial Safety at Mohamed Ben Ahmed University of Oran 2, and director of the Laboratory of Industrial Safety and Sustainable Development Engineering (LISIDD). 

Previously, she obtained her PhD at the University of Oran, within the LIO and the LAMIH (UMR 8201). 

Her research focuses in particular on improving the performance of industrial production systems through machine learning, as well as on the reliability and safety of industrial and energy systems.