Publications
Chosen filters:
Chosen filters:
Rotating machinery speed extraction through smartphone video acquisition from a radial viewpoint KU Leuven
This article investigates the viability of using a smartphone as a cost-effective tool for rotational speed extraction. The proposed method exploits the geometrical image deformations induced by a smartphone’s rolling shutter camera, which reads pixel lines sequentially with a delay equal to the rolling shutter period. This characteristic period is the basis for the developed methodology and allows measuring constant and varying speeds above the ...
Predicting pitting severity in gearboxes under unseen operating conditions and fault severities using convolutional neural networks with power spectral density inputs KU Leuven
The PHM North America 2023 Data Challenge tasked participants to diagnose the pitting fault severity of a gearbox from a three-channel vibration signal. This work summarizes the authors’ proposed diagnostics solution which consists of a convolutional neural network with an ordinal loss criterion, trained on the power spectral density of the signal. This method is selected based on a rigorous evaluation using three dedicated validation sets, ...
Differentiable Short-Time Fourier Transform Window Length Selection Driven by Cyclo-Stationarity KU Leuven
Cavitation detection based on instantaneous angular speed KU Leuven
The study and detection of cavitation in hydraulic turbo-machinery has gained recently particular attention from both manufacturers and end users, due to the damage caused on machines’ surfaces that may result in costly sudden shutdowns. The traditional cavitation detection approaches exploit the noise and vibration signals that sense the vapor bubble implosion pressure waves, while advanced demodulation techniques are used in order to locate ...
Condition Monitoring of Wind Turbines Based on Anomaly Detection Using Deep Support Vector Data Description KU Leuven
Wind turbine condition monitoring is considered a key task in the wind power industry. A plethora of methodologies based on machine learning have been proposed for monitoring wind turbines, but the absence of faulty data at the amount and the variety needed still set limitations. Therefore, anomaly detection (AD) methodologies are proposed as alternatives for fault detection. Deep learning tools have been introduced in the research field of wind ...