Postdoctoral Researcher in Structural Health Monitoring (SHM) and Embedded Predictive Maintenance

Institut Mines-Télécom; Douai; Sur place; EUR 35 000 - 45 000; Plein temps

Résumé du poste

A prestigious research institution in France is seeking a Postdoctoral Researcher in Structural Health Monitoring and Embedded Predictive Maintenance. The successful candidate will have a Ph.D. and a strong background in machine learning and signal processing. Responsibilities include designing data-analysis methods and implementing advanced machine-learning models. The position is on-site in Douai with a duration of 18 months, starting in early 2026. Applicants should be proficient in Python, with experience in interpreting experimental data.

Qualifications

• Ph.D. in relevant field.
• Strong background in machine learning and deep learning.
• Experience in signal processing and data interpretation.
• Proficiency in Python and scientific libraries.
• Strong analytical skills and self-initiative.

Responsabilités

• Design and validate data-analysis methods for damage detection and fault diagnosis.
• Perform advanced signal and time-series analysis.
• Implement machine-learning models for fault diagnosis and prognostics.
• Integrate models into predictive-maintenance frameworks.
• Contribute to scientific dissemination and project deliverables.

Connaissances

Machine learning

Deep learning

Time-series analysis

Signal processing

Python

Acoustic-emission analysis

Data interpretation

Formation

Ph.D. in Computer Science or related field

Outils

PyTorch

TensorFlow

scikit-learn

NumPy

Description du poste

Postdoctoral Researcher in Structural Health Monitoring (SHM) and Embedded Predictive Maintenance

• On-site

Host Unit: CERI Digital Systems

Duration: 18 months

Expected Start Date: February or March 2026

Context: Public establishment belonging to IMT (Institut Mines‑Télécom), under the supervision of the Ministry of Economy, Finance and Digital Sovereignty. IMT Nord Europe aims to provide students with ethically responsible engineering practice, conduct high‑impact R&D and support territorial development. Ideally positioned at the heart of Europe—1h from Paris, 30min from Brussels, 1h30 from London—it has strong ambitions to be a key actor in industrial transitions, digital and environmental, combining education and research.

Location: Douai and Lille campuses, research facilities covering ~20,000 m².

The Digital Systems centre bridges the physical and digital worlds by modelling and optimizing complex systems, enhancing human–machine interactions, and designing secure, connected systems at all Technology Readiness Levels.

Project: Maghydro

The Maghydro project seeks to enhance safety, reliability, and lifetime prediction of pressurized hydrogen storage systems through advanced monitoring and data‑driven approaches. It combines experimental testing, damage characterization, and predictive modeling to develop embedded SHM and predictive maintenance solutions.

Key Objectives

• Pressure testing (static and fatigue) of instrumented composite bottles equipped with strain gauges, accelerometers, and acoustic emission (AE) sensors.
• Characterisation of damage mechanisms via acoustic emission, strain damage and accelerometer analysis.
• Assessment and classification of defects, including manufacturing defect criticality and progressive damage evolution.
• Health monitoring of in‑service bottles, leading to embedded predictive maintenance systems.

Responsibilities

• Design and validate data‑analysis and machine‑learning methods for damage detection, fault diagnosis, and health‑indicator estimation using lab and in‑service data.
• Perform advanced signal and time‑series analysis to detect anomalies, characterise acoustic‑emission events, and estimate structural health indicators.
• Implement and evaluate machine‑learning and deep‑learning models, including weakly or semi‑supervised approaches, for fault diagnosis, defect classification, prognostics and Remaining Useful Life (RUL) estimation.
• Integrate models into an embedded predictive‑maintenance framework that ensures real‑time applicability and robustness to environmental variability.
• Contribute to scientific dissemination and project deliverables: write papers and reports, present results at meetings and conferences, collaborate with experimental and modelling teams within the consortium.

Qualifications

• Ph.D. in Computer Science, Applied Mathematics, Mechanics, Control Engineering, or related field.
• Strong background in machine‑learning, deep‑learning, and time‑series analysis.
• Experience in signal processing, acoustic‑emission analysis, or sensor‑data interpretation.
• Proficiency in Python and scientific libraries (PyTorch/TensorFlow, scikit‑learn, NumPy, etc.).
• Ability to handle and interpret experimental data from multi‑sensor systems.
• Familiarity with SHM or predictive‑maintenance concepts is highly desirable.
• Scientific curiosity, autonomy, initiative, strong analytical and problem‑solving skills.
• Excellent written and oral communication in English (French is a plus).

Application

Information and applications: Dr. Lala Rajaoarisoa, Lecturer and Researcher – e‑mail: [e-mail]; Tel.: [phone].

This position is offered to civil servants on a mobility basis, or under public‑law contract. The role can be adapted for a disabled person.

Deadline for submissions: 10/01/2026.