Machine condition monitoring using principal component representations

He, Qingbo; Yan, Ruqiang; Kong, Fanrang; Du, Ruxu

doi:10.1016/j.ymssp.2008.03.010

Cited by 91 publications

(48 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In another study (14) , PCA was used to extract the lowdimensional Principal Component Representations from the statistical features of the measured signals to monitor machine conditions. Several techniques have been used for early fault detection and health monitoring of rotating machines.…”

Section: Introductionmentioning

confidence: 99%

Technique for optimal placement of transducers for fault detection in rotating machines

Fatima

Dastidar

Mohanty

et al. 2013

Proceedings of the Institution of Mechanical Engineers, Part O:

View full text Add to dashboard Cite

Online fault detection and diagnosis of rotating machinery requires a number of transducers which can be significantly expensive for industrial processes. The sensitivity of various transducers and their appropriate positioning are dependent on different types of fault conditions. It is critical to formulate a method to systematically determine the effectiveness of transducer locations for monitoring the condition of a machine. In this paper, Number of Independent Sources analysis is used as an effective tool for reducing the number of vibration sources within the system which is then followed by Principal Component Analysis to identify the incoherent transducers to be employed for fault detection. This experiment is conducted on a machine fault simulator for unbalanced rotor, 1 misaligned shaft and cracked shaft. The validation of the proposed selection process is illustrated using spectral analysis for each defect.

show abstract

Section: Introductionmentioning

confidence: 99%

Technique for optimal placement of transducers for fault detection in rotating machines

Fatima

Dastidar

Mohanty

et al. 2013

Proceedings of the Institution of Mechanical Engineers, Part O:

View full text Add to dashboard Cite

show abstract

“…PCA is a simple nonparametric method which can extract the most relevant information from a set of redundant or noisy data and form some new variables, the principal components, and explained the maximum amount of variability of the original data. In the area of machine condition monitoring, PCA method has been investigated to identify the most representative features from a variety of characteristic features of roller bearings and gearbox in time, frequency and or time-frequency domains [14,15]. The effectiveness of PCA has been verified experimentally on a bearing test machine, the results validated the suitability of the PCA features selection scheme [14].…”

Section: Introductionmentioning

confidence: 91%

Machine Tool Volumetric Error Features Extraction and Classification Using Principal Component Analysis and K-Means

Xing

Mayer

Achiche

2018

JMMP

View full text Add to dashboard Cite

Volumetric errors (VE) are related to the machine tool accuracy state. Extracting features from the complex VE data provides with a means to characterize this data. VE feature classification can reveal the machine tool accuracy states. This paper presents a study on how to use principal component analysis (PCA) to extract the features of VE and how to use the K-means method for machine tool accuracy state classification. The proposed data processing methods have been tested with the VE data acquired from a five-axis machine tool with different states of malfunction. The results indicate that the PCA and K-means are capable of extracting the VE feature information and classifying the fault states including the C axis encoder fault, uncalibrated C axis encoder fault, and pallet location fault from the machine tool normal states. This research provides a new way for VE features extraction and classification.

show abstract

“…As has been shown in Clifton et al (2006), this step has a major effect for the novelty detection system since it enables a better discriminating capability between the two different classes. Scaling and normalization is also important for most condition monitoring systems for the removal of any undesirable environmental or operational effects in the analyzed data (He et al, 2009). As a preprocessing method, it is considered for improving the performance of one-class classifiers (Juszczak et al, 2002): it is a very good practise when working with machinelearning algorithms to scale the data being analyzed, since large absolute value ranges of features will tend to dominate the ones with smaller value ranges (Hsu et al, 2016).…”

Section: Preprocessing Of Raw Vibration Datamentioning

confidence: 99%

Vibration Monitoring of Gas Turbine Engines: Machine-Learning Approaches and Their Challenges

Matthaiou

Khandelwal

Antoniadou

2017

Front. Built Environ.

View full text Add to dashboard Cite

In this study, condition monitoring strategies are examined for gas turbine engines using vibration data. The focus is on data-driven approaches, for this reason a novelty detection framework is considered for the development of reliable data-driven models that can describe the underlying relationships of the processes taking place during an engine's operation. From a data analysis perspective, the high dimensionality of features extracted and the data complexity are two problems that need to be dealt with throughout analyses of this type. The latter refers to the fact that the healthy engine state data can be non-stationary. To address this, the implementation of the wavelet transform is examined to get a set of features from vibration signals that describe the non-stationary parts. The problem of high dimensionality of the features is addressed by "compressing" them using the kernel principal component analysis so that more meaningful, lowerdimensional features can be used to train the pattern recognition algorithms. For feature discrimination, a novelty detection scheme that is based on the one-class support vector machine (OCSVM) algorithm is chosen for investigation. The main advantage, when compared to other pattern recognition algorithms, is that the learning problem is being cast as a quadratic program. The developed condition monitoring strategy can be applied for detecting excessive vibration levels that can lead to engine component failure. Here, we demonstrate its performance on vibration data from an experimental gas turbine engine operating on different conditions. Engine vibration data that are designated as belonging to the engine's "normal" condition correspond to fuels and airto-fuel ratio combinations, in which the engine experienced low levels of vibration. Results demonstrate that such novelty detection schemes can achieve a satisfactory validation accuracy through appropriate selection of two parameters of the OCSVM, the kernel width γ and optimization penalty parameter ν. This selection was made by searching along a fixed grid space of values and choosing the combination that provided the highest cross-validation accuracy. Nevertheless, there exist challenges that are discussed along with suggestions for future work that can be used to enhance similar novelty detection schemes.

show abstract

Machine condition monitoring using principal component representations

Cited by 91 publications

References 17 publications

Technique for optimal placement of transducers for fault detection in rotating machines

Technique for optimal placement of transducers for fault detection in rotating machines

Machine Tool Volumetric Error Features Extraction and Classification Using Principal Component Analysis and K-Means

Vibration Monitoring of Gas Turbine Engines: Machine-Learning Approaches and Their Challenges

Contact Info

Product

Resources

About