Improving the performance of univariate control charts for abnormal detection and classification
详细信息 查看全文
- 作者：Christos Yiakopoulosa ; ; Maria Koutsoudakia ; Konstantinos Grylliasb ; c ; Ioannis Antoniadisa
- 关键词：SNR ; Signal-Noise Ratio ; SVD ; Singular Value Decomposition ; SV ; Singular Value ; FC ; Center Frequency ; CSMW ; Complex Shifted Morlet wavelet ; EWMA ; Exponentially Weight Moving Average ; MSPC ; Multivariate Statistical Process Control ; MM ; Mathematical Morphology ; MO ; Morphological Operator ; BG ; Beucher Gradient ; SF ; Shape Factor ; KU ; Kurtosis ; RMS ; Root Mean Square ; CF ; Crest Factor ; SMF ; Shape Modified Form ; ESPRIT ; Estimation of Signal Parameters via Rotational Invariance Techniques ; ENKU ; ENvelop
- 刊名：Mechanical Systems and Signal Processing
- 出版年：2017
- 出版时间：1 March 2017
- 年：2017
- 卷：86
- 期：part_PA
- 页码：122-150
- 全文大小：2867 K
文摘
- •
Beucher gradient and CSMW are used to extract singular values as key features.
- •
The method can improve the anomaly detection performance of the control charts.
- •
The method not only successfully detects anomalies, but also classifies the faults.
- •
The SORDINA technique is applied to detect an actual source frequency.
- •
IFESIS combines the advantages of the CSMW and the subspace based approaches.