Seminar by Douw Marx and Xian Wu

Joint seminar : Signal processing for fault detection and acoustic monitoring

Xian Wu is a PhD student at KU Leuven in Belgium, working under the supervision of Professor Konstantinos Gryllias. Xian completed their Master's degree in Mechanical Engineering at RWTH Aachen University in Germany and gained further experience at HEAD acoustics, where they developed expertise in acoustics and machine learning.
Xian's current research interests focus on acoustic monitoring and localization using sparse microphone arrays and beamforming techniques. He’s working on two main projects: the first project aims to develop a non-contact solution for early detection of bearing failures in industrial environments, leveraging acoustic imaging and a sparse microphone array to enhance signal-to-noise ratios of rolling element bearing signatures.
The second project explores the integration of acoustic localization and augmented reality to assist pedestrians in avoiding critical traffic encounters involving nearby moving vehicles. This system uses a microphone array to detect out-of-view vehicles and displays them through augmented reality visualization, enhancing pedestrian safety.
Currently, Xian's acoustic monitoring research is the subject of the Marie Curie ECO-DRIVE research secondment at INSA Lyon, under the supervision of Prof. Jerome Antoni.

Douw Marx is a PhD student at KU Leuven in Belgium, working under the supervision of Professor Konstantinos Gryllias. He completed his Master's thesis with a focus on condition-based maintenance at the University of Pretoria in South Africa and subsequently worked as a data analyst in the pyrometallurgy industry.
Douw's current research interests include improving fault detection models for rotating machinery by integrating engineering domain knowledge. He is pursuing two main approaches: the first involves applying domain-knowledge-based regularizations to unsupervised data-driven models, while the second leverages advances in automatic differentiation to create differentiable signal processing architectures using deep learning frameworks.
This second approach is also the current subject of his Marie Curie MOIRA research secondment at INSA Lyon under the supervision of Prof. Jerome Antoni.