My research focuses on the intersection of Artificial Intelligence, Computer Vision, and Electrical Engineering, with a primary emphasis on Digital Twin technology for marine renewable energy and industrial fault detection.

Current Research (PhD)

TWIST Project: AI-based Digital Twin Technology

Institute: Université de Bretagne Occidentale (UBO)
Location: Brest, France
Duration: October 2025 – October 2028
Role: Doctoral Researcher

  • Developing an AI-driven Digital Twin framework for the diagnosis and prognosis of tidal turbine failures.
  • Investigating multi-modal approaches to detect and predict biofouling and mechanical faults.
  • Applying Machine Learning, Deep Learning, and Signal Processing to improve the reliability of Marine Renewable Energy (MRE) systems.
  • Utilizing Computer Vision techniques for structural health monitoring and underwater image analysis.

Submitted Publications

  • Review Paper: “Digital Twin-Driven Multimodal Intelligence for Fault Diagnosis and Biofouling Monitoring in Tidal Stream Turbines: A Comprehensive Review”. This work investigates physics-based models, signal processing, and emerging Large Language Model (LLM)-assisted diagnosis frameworks. Submitted to the journal Ocean Engineering.

Conferences & Research Activities

  • IECON 2026 – 52nd Annual IEEE Industrial Electronics Conference: Accepted conference paper titled “Transformer-based Biofouling Detection on Tidal Stream Turbines Under Severe Data Scarcity”, co-authored with Y. Belkhier, H. Rashid, and M. Benbouzid. The paper presents a Transformer-based approach for biofouling detection on tidal stream turbines under severe data scarcity.
  • 2026 INORE North American Symposium: Participated in the symposium held in Hoboken, New Jersey, USA, where I presented my research through a poster presentation and seminar talk focused on AI-driven Digital Twin technologies for tidal stream turbines, with particular emphasis on biofouling detection, failure diagnosis, and prognosis. The symposium also provided valuable opportunities for technical discussions, research exchange, international collaboration, and visits to leading marine renewable energy research laboratories and industry facilities.

Past Research Contributions

Deepfake Detection (Master's Thesis)

  • Developed a diagnostic system to identify deepfakes using lighting inconsistencies in retinal and facial movements.
  • Implemented advanced landmark detection algorithms for biometric verification.
  • Utilized FaceSwapGAN architectures to establish a baseline for generative adversarial detection.

Computer Vision-based Activity Recognition

  • Researched and developed an automated system for activity recognition on construction sites.
  • Integrated CNN and LSTM architectures to capture both spatial features and temporal sequences in video data.

Industrial AI: Automated Billing & Steganography

  • YOLOv5 Billing System: Developed an AI-based automated stationery calculator using YOLOv5 for real-time item detection.
  • Multi-Image Steganography: Implemented a deep learning framework for the enhancement of encoded covers, focusing on data security and visual fidelity.

Research Interests

  • Digital Twins: Real-time modeling for predictive maintenance and industrial monitoring.
  • Renewable Energy AI: Optimization and fault diagnosis for tidal and offshore wind turbines.
  • Computer Vision: Object detection (YOLO), biometric analysis, and underwater imaging.
  • Deep Learning: CNNs, LSTMs, and GANs for complex signal and image processing.
  • Signal Processing: Analyzing sensor data for industrial prognosis and health management (PHM).

Research Skills

  • Artificial Intelligence: Deep Learning (PyTorch/TensorFlow), Machine Learning (Scikit-learn), and RAG-based systems.
  • Computer Vision: YOLOv8/v11, OpenCV, Image Segmentation, and Landmark Detection.
  • Technical Computing: Python, MATLAB, Signal Processing, and Data Preprocessing.
  • Academic Writing: LaTeX, Overleaf, BibTeX, and Research Methodology.

Research Goal

My goal is to bridge the gap between AI research and industrial application. By developing advanced Digital Twins, I aim to provide scalable, intelligent solutions for monitoring complex systems—specifically within the marine renewable energy sector—to ensure sustainable and reliable energy production.