21-11-2024 - Lecture Hall E

PhD presentation

Predictive Maintenance of Marine Diesel Engines: Integrating In-Cylinder Pressure Measurements, Machine Learning, and First-Principles Models

Youri Linden

This study addresses the pressing need for advanced maintenance strategies to enhance operational readiness of critical assets, particularly Marine Diesel Engines in naval vessels. Against a backdrop of heightened geopolitical tensions and increased defense budgets, optimizing maintenance through data-driven and condition-based approaches has become essential. The research focuses on leveraging advanced condition monitoring techniques, such as in-cylinder pressure measurements, to detect early signs of engine degradation. A key objective is to adapt these methods to real-world operational settings characterized by noisy, incomplete data and environmental variability. By bridging gaps in existing literature, the study aims to develop hybrid models integrating machine learning and physical principles for real-time fault detection and degradation prediction. With access to both controlled experimental and operational naval data, this work seeks to deliver robust, practical solutions, enhancing the predictive maintenance capabilities essential for naval operational efficiency and asset reliability.

21-11-2024 - Lecture Hall E

PhD presentation

Deep sea mining turbidity current dispersion on a slope

Shaheen Wahab

The global transition to clean energy technologies, critical for achieving a low-carbon future, intensifies the demand for essential raw materials such as nickel, cobalt, and rare earth metals. Land-based mining faces challenges like ore grade depletion and environmental concerns, prompting exploration of alternative sources like deep-sea mining (DSM). DSM, particularly in regions like the Clarion-Clipperton Zone (CCZ), offers access to critical metals in polymetallic nodules but raises environmental concerns, notably sediment plumes that threaten deep-sea ecosystems. This research investigates the near-field dispersion dynamics of turbidity currents from DSM activities, focusing on slope angle and velocity effects. Experiments conducted in TU Delft’s Offshore and Dredging Laboratory used a flume tank setup, simulating turbidity currents with glass beads as sediment. Various measurement techniques such as Acoustic Doppler velocimeters, ultrasonic velocity profilers, and multi-angle cameras revealed that steeper slopes and higher velocity ratios increased sediment bulge formation and dispersion. Downhill experiments exhibited wider dispersion angles with higher velocity ratios, while uphill conditions showed inverse trends. Findings emphasize slope angle and discharge velocity as key factors influencing sediment plume behaviour.