Yacine Belhadj

Electromechanical engineer with a PhD in Engineering (highest distinction), specializing in machine learning, deep learning and signal processing. Experienced in developing and deploying data-driven solutions for anomaly detection, predictive maintenance and industrial monitoring. Strong background in Python, model development and evaluation, with experience applying machine learning to large-scale industrial datasets.

Proven ability to translate complex research into practical solutions through collaboration with industrial partners. Currently leading a work package in a collaborative R&D project, coordinating technical activities, research planning and deliverables.

Postdoctoral Researcher at OWI-Lab, Vrije Universiteit Brussel (2026-05 – Present)

• Lead a work package in a collaborative R&D project, managing technical objectives, research planning, task coordination and deliverables while developing machine learning methods for model drift detection, transfer learning and uncertainty quantification.
• Developed a machine learning solution for fatigue assessment across two Belgian offshore wind farms, outperforming previous methods while requiring less input data. The solution supports more accurate fatigue assessments and informed decisions on extending wind turbine operational lifetimes by an additional two years.
• Processed and analyzed 11 TB of high-frequency sensor data from an offshore power cable, developing scalable signal processing workflows for automated event detection, classification and anomaly identification, including the assessment of potential cable unburial.
• Designed and delivered a hands-on workshop on applying deep learning to structural monitoring.
• Developed an audio-classification with domain adaptation approach for the international DCASE 2026 machine-learning challenge, placing seventh in the competition.

Researcher in Machine Learning and Structural Health Monitoring at OWI-Lab, Vrije Universiteit Brussel (2021 – 2026)

• Developed and deployed machine learning models to detect structural damage and assess operational conditions across offshore wind turbines, railway bridges, electricity transmission towers and pedestrian bridges, combining engineering expertise with data-driven methods to identify abnormal behavior and changes in structural performance.
• First author of three papers in leading peer-reviewed (Q1) journals on machine learning for structural monitoring. Contributed to conference papers presented at eight European conferences and one international conference at Stanford.
• Collaborated with the German research institute FKIE on an international challenge focused on detecting abnormal machine sounds.
• Taught Mechanical Vibrations laboratory sessions and created Python-based exercises with automated assessment to support student learning.

Data Analyst Intern at 24SEA (2020 – 2020)

Summer 2020

• Analyzed sensor data to investigate sensor failures and identify abnormal measurements.

Automotive Workshop Assistant (2015 – 2015)

Summer 2015

• Assisted with engine repairs and used a lathe to repair engine components.

Sales Assistant at Bel-TV (2014 – 2014)

• Supporting sales and customer service.

PhD in Engineering – Vrije Universiteit Brussel, Belgium (2021 – 2026)

MSc in Electromechanical Engineering, Energy – École polytechnique de Bruxelles, ULB (2019 – 2021)

BSc in Electromechanical Engineering – École polytechnique de Bruxelles, ULB (2016 – 2019)