Health Index Estimation and RUL Prediction on Gear Pumps
Master thesis
Permanent lenke
https://hdl.handle.net/11250/3089209Utgivelsesdato
2023Metadata
Vis full innførselSamlinger
- Studentoppgaver (TN-IDE) [823]
Beskrivelse
Full text not available
Sammendrag
This thesis explores innovative machine learning and signal processing techniques toaccurately estimate the wear by constructing Health Indexes and predict the RemainingUseful Life (RUL) of gear pumps used in ABB Robotics’ car painting robots. The studyinvestigates pressure sensor data obtained from the gear pumps and identifies changes inthe data as wear progresses.
Two methods are proposed for Health Index estimation: One leveraging Fast FourierTransform (FFT) values in a weighted sum of the significant frequencies and anotherutilizing Convolutional Neural Network Autoencoders (CNN-AE) supplemented with anoutlier detection model (One Class SVM / Isolation Forest). The constructed Healthindexes for both methods successfully represent the degradation observed on the pumps.
The proposed RUL models utilize these constructed Health Indexes to make predictions,resulting in improved predictions compared to using only the current lifetime as input.Using the feature vector from a CNN-AE network supplied with FFT values as inputto a One-Class SVM model for Health Index estimation and then utilizing these in aBidirectional LSTM model yields the best RUL predictions. However, the other proposedmethods are also performing satisfactorily.
This study contributes to the field of predictive maintenance for gear pumps and demonstrate the effectiveness of data-driven techniques in estimating wear and predicting RULin this context.