A
Anna Bartkowiak
Researcher at University of Wrocław
Publications - 63
Citations - 432
Anna Bartkowiak is an academic researcher from University of Wrocław. The author has contributed to research in topics: Principal component analysis & Set (abstract data type). The author has an hindex of 9, co-authored 61 publications receiving 401 citations.
Papers
More filters
Journal ArticleDOI
Two simple multivariate procedures for monitoring planetary gearboxes in non-stationary operating conditions
Radoslaw Zimroz,Anna Bartkowiak +1 more
TL;DR: A novel way for condition monitoring of planetary gearboxes based on multivariate statistics is suggested and the emphasis is put on the algebraic and geometric interpretations of the PCA.
Journal ArticleDOI
Investigation on Spectral Structure of Gearbox Vibration Signals by Principal Component Analysis for Condition Monitoring Purposes
Radoslaw Zimroz,Anna Bartkowiak +1 more
TL;DR: In this paper, a case study for planetary gearboxes in good and bad conditions (case B and case A) was provided, where 15 amplitudes of spectral components related to fundamental planetary mesh frequency and its harmonics were used as diagnostic features.
Journal ArticleDOI
Outliers analysis and one class classification approach for planetary gearbox diagnosis
Anna Bartkowiak,Radoslaw Zimroz +1 more
TL;DR: An application of outlier analysis for diagnosis of gearboxes working under non-stationary operating conditions is considered, using quite large data characterizing two gearboxes, one in bad and the other in good condition.
Journal ArticleDOI
Dimensionality reduction via variables selection – Linear and nonlinear approaches with application to vibration-based condition monitoring of planetary gearbox
Anna Bartkowiak,Radoslaw Zimroz +1 more
TL;DR: A novel hybrid approach is applied: all subset search by using multivariate linear regression (MLR) and variables shrinkage by the least absolute selection and shrinkage operator (Lasso) performing a non-linear approach, which gave consistent results and yielded subsets with healthy or faulty diagnostic properties.
Journal ArticleDOI
Novel method of informative frequency band selection for vibration signal using Nonnegative Matrix Factorization of spectrogram matrix
TL;DR: A novel method of informative frequency band selection is proposed that utilizes the approach of Non-negative Matrix Factorization applied to time-frequency signal representation to filter particular components out of the original signal.