A novel health indicator developed using filter-based feature selection algorithm for the identification of rotor defects

2020 ◽  
pp. 1748006X2091695 ◽  
Author(s):  
Anil Kumar ◽  
CP Gandhi ◽  
Xiaoyang Liu ◽  
Yi Liu ◽  
Yuqing Zhou ◽  
...  
Keyword(s):  
Feature Selection ◽  
Selection Process ◽  
Signal Frequency ◽  
Health Indicator ◽  
Angular Misalignment ◽  
Time Frequency ◽  
Vibration Data ◽  
Time Domain Signal ◽  
Frequency Representation ◽  
Parallel Misalignment

In this work, a novel health indicator is developed for the identification of rotor defects. The indicator is developed by extracting features from vibration data acquired from horizontal and vertical directions of rotors. A total of 38 features were initially extracted from time-domain signal, frequency-domain signal, and time–frequency representation. Out of many features, six most important features were selected using filter-based feature selection process. Thereafter, important features were fused together using manifold learning to develop health indicator. The developed indicator is used to identify misalignments (angular misalignment and parallel misalignment), rub, and unbalance. The major benefit of the proposed method is that it not only indicates the presence of defect in the rotor but also indicates the severity of defect. The experimental study presented in this article justifies that the proposed method is sensitive to the increasing levels of horizontal and angular misalignment and unbalance. The developed indicator is sensitive enough to indicate the presence of rub.

scholarly journals Intelligent Deep Adversarial Network Fault Diagnosis Method Using Semisupervised Learning

2020 ◽  
Vol 2020 ◽  
pp. 1-13
Author(s):  
Juan Xu ◽  
Yongfang Shi ◽  
Lei Shi ◽  
Zihui Ren ◽  
Yang Lu
Keyword(s):  
Fault Diagnosis ◽  
Original Data ◽  
Semisupervised Learning ◽  
Learning Ability ◽  
Time Frequency ◽  
Vibration Data ◽  
Time Domain Signal ◽  
Adversarial Network ◽  
Vibration Time ◽  
Network Fault

In recent years, deep learning has become a popular issue in the intelligent fault diagnosis of industrial equipment. Under practical working conditions, although the collected vibration data are of large capacity, most of the vibration data are not labeled. Collecting and labeling sufficient fault data for each condition are unrealistic. Therefore, constructing a reliable fault diagnosis model with a small amount of labeled vibration data is a significant problem. In this paper, the vibration time-domain signal of the fault bearing is transformed into a 2-dimensional image by wavelet transform to obtain the time-frequency domain information of the original data. A deep adversarial convolutional neural network based on semisupervised learning is proposed. A large amount of fake data generated by the generator and unlabeled true vibration data are used in the discriminator to learn the overall distribution of data by judging the authenticity of the input. Three regular terms for different loss functions are designed to constrain the parameters of the discriminator to improve the learning ability of the model. The proposed method is validated by two bearing fault diagnosis cases. The experiment results show that the proposed method has higher diagnostic accuracy than traditional deep models on multigroup small datasets of different capacities. The proposed method provides a new solution to the fault diagnosis problem with large vibration data but few labels.

Two-Stage Adaptive Line Enhancer and Sliced Wigner Trispectrum for the Characterization of Faults from Gear Box Vibration Data

1999 ◽  
Vol 121 (4) ◽  
pp. 488-494 ◽  
Author(s):  
S. K. Lee ◽  
P. R. White
Keyword(s):  
Background Noise ◽  
Two Stage ◽  
Vibration Signals ◽  
Time Frequency ◽  
Vibration Data ◽  
Frequency Representation ◽  
Frequency Domains ◽  
Adaptive Line Enhancer ◽  
Gear Fault

Impulsive sound and vibration signals in gears are often associated with faults which result from impacting and as such these impulsive signals can be used as indicators of faults. However it is often difficult to make objective measurements of impulsive signals because of background noise signals. In order to ease the measurement of impulsive sounds embedded in background noise, it is proposed that the impulsive signals are enhanced, via a two stage ALE (Adaptive Line Enhancer), and that these enhanced signals are then analyzed in the time and frequency domains using a Wigner higher order time-frequency representation. The effectiveness of this technique is demonstrated by application to gear fault data.

HILBERT SPECTRUM IN TIME-FREQUENCY REPRESENTATION OF AUDIO SIGNALS CONSIDERING DISJOINT ORTHOGONALITY

2010 ◽  
Vol 02 (03) ◽  
pp. 313-336 ◽  
Author(s):  
MD. KHADEMUL ISLAM MOLLA ◽  
KEIKICHI HIROSE
Keyword(s):  
Hilbert Transform ◽  
Time Domain ◽  
Adaptive Method ◽  
Intrinsic Mode Functions ◽  
Audio Signals ◽  
Hilbert Spectrum ◽  
Time Frequency ◽  
Time Domain Signal ◽  
Frequency Representation ◽  
Non Stationary Signal

The performance of Hilbert spectrum (HS) in time-frequency representation (TFR) of audio signals is investigated in this paper. HS offers a fine-resolution TFR of time domain signals. It is derived by applying empirical mode decomposition (EMD), a newly developed data adaptive method for nonlinear and non-stationary signal analysis together with Hilbert transform. EMD represents any time domain signal as a sum of a finite number of bases called intrinsic mode functions (IMFs). The instantaneous frequency responses of the IMFs derived through Hilbert transform are arranged to obtain the TFR of the analyzing signal yielding the HS. The disjoint orthogonal property of audio signals is used as the decisive factor to measure the efficiency in TFR. Several audio signals are considered as disjoint orthogonal if not more than one source is active at any time-frequency cell. The performance of HS is compared with well known and widely used short-time Fourier transform technique for TFR. The experimental results show that HS based method performs better in time-frequency representation of the audio signals with the consideration of disjoint orthogonality.

A review of time–frequency representations, with application to sound/music analysis–resynthesis

1997 ◽  
Vol 2 (3) ◽  
pp. 193-205 ◽  
Author(s):  
PAUL MASRI ◽  
ANDREW BATEMAN ◽  
NISHAN CANAGARAJAH
Keyword(s):  
Music Analysis ◽  
Intermediate Representation ◽  
Short Time Fourier Transform ◽  
Time Frequency ◽  
Time Domain Signal ◽  
Analysis Process ◽  
Frequency Representation ◽  
The Time Domain ◽  
Higher Order Spectra ◽  
Short Time

Analysis–resynthesis (A–R) systems gain their flexibility for creative transformation of sound by representing sound as a set of musically useful features. The analysis process extracts these features from the time domain signal by means of a time–frequency representation (TFR). The TFR provides an intermediate representation of sound that must make the features accessible and measurable to the rest of the analysis. Until very recently, the short-time Fourier transform (STFT) has been the obvious choice for time–frequency representation, despite its limitations in terms of resolution. Recent and ongoing developments are providing several alternative schemes that allow for a more considered choice of TFR. This paper reviews these contemporary approaches in comparison with the more classical ones and with reference to their applicability, merits and shortcomings for application to sound analysis. (Where they have been successfully applied, details are provided.) The techniques reviewed include linear, bilinear and higher-order spectra, nonparametric and parametric methods and some sound-model-specific TFRs.

The importance of the time–frequency representation for sound/music analysis–resynthesis

1997 ◽  
Vol 2 (3) ◽  
pp. 207-214
Author(s):  
PAUL MASRI ◽  
ANDREW BATEMAN ◽  
NISHAN CANAGARAJAH
Keyword(s):  
System Design ◽  
Time Domain ◽  
Music Analysis ◽  
Short Time Fourier Transform ◽  
Time Frequency ◽  
Time Domain Signal ◽  
Frequency Representation ◽  
Initial Stage ◽  
Original Time ◽  
Short Time

The time–frequency representation (TFR) is the initial stage of analysis in sound/music analysis–resynthesis (A–R) systems. Given a time-domain waveform, the TFR makes temporal and spectral detail available to the remainder of the analysis, so that the component features may be extracted. The resulting ‘feature set’ must represent the sound as completely as the original time-domain signal, if the A–R system is to be capable of effective transformation and good synthesis sound quality. Therefore the system as a whole is reliant upon the TFR to make the sound components detectable, separable and measurable. Yet the standard TFR to-date is the short-time Fourier transform (STFT), of which the shortcomings, in terms of resolution, are well recognised. The purpose of this paper is to demonstrate the importance of the TFR to system function and system design. Poor feature extraction is shown to result from the use of inappropriate TFRs, whose underlying assumptions and expectations do not match those of the system. Existing models are used as case studies, with examples of performance for different sound types. A philosophy for A–R system design that includes TFR design is presented and a methodology for implementing it is proposed.

Detecting Cracks in Reciprocating Compressor Valves Using Pattern Recognition in Frequency Space

2012 ◽  
Author(s):  
Kurt Pichler ◽  
Edwin Lughofer ◽  
Thomas Buchegger ◽  
Erich Peter Klement ◽  
Matthias Huschenbett
Keyword(s):  
Pattern Recognition ◽  
Measurement Data ◽  
Main Idea ◽  
Reciprocating Compressor ◽  
Frequency Space ◽  
Time Frequency ◽  
Vibration Data ◽  
Frequency Representation ◽  
Data Driven Approach ◽  
Fault State

This paper presents a novel data-driven approach for detecting cracks in reciprocating compressor valves by analyzing vibration data. The main idea is that the time-frequency representation will show typical patterns, depending on the fault state and other variables. The problem of detecting these patterns reliably is solved by taking a detour via two dimensional autocorrelation. This emphasizes the patterns and reduces noise effects, thus identifying appropriate features becomes easier. The features are then classified using well known pattern recognition approaches. The methods performance is validated by analyzing real world measurement data.

scholarly journals Strain Sensitivity Enhancement of Broadband Ultrasonic Signals in Plates Using Spectral Phase Filtering

2021 ◽  
Vol 11 (6) ◽  
pp. 2582
Author(s):  
Lucas M. Martinho ◽  
Alan C. Kubrusly ◽  
Nicolás Pérez ◽  
Jean Pierre von der Weid
Keyword(s):  
Fourier Transform ◽  
Guided Waves ◽  
Strain Level ◽  
Energy Concentration ◽  
Short Time Fourier Transform ◽  
Time Frequency ◽  
Frequency Representation ◽  
The Fourier Transform ◽  
Short Time ◽  
Sensitivity Gain

The focused signal obtained by the time-reversal or the cross-correlation techniques of ultrasonic guided waves in plates changes when the medium is subject to strain, which can be used to monitor the medium strain level. In this paper, the sensitivity to strain of cross-correlated signals is enhanced by a post-processing filtering procedure aiming to preserve only strain-sensitive spectrum components. Two different strategies were adopted, based on the phase of either the Fourier transform or the short-time Fourier transform. Both use prior knowledge of the system impulse response at some strain level. The technique was evaluated in an aluminum plate, effectively providing up to twice higher sensitivity to strain. The sensitivity increase depends on a phase threshold parameter used in the filtering process. Its performance was assessed based on the sensitivity gain, the loss of energy concentration capability, and the value of the foreknown strain. Signals synthesized with the time–frequency representation, through the short-time Fourier transform, provided a better tradeoff between sensitivity gain and loss of energy concentration.

scholarly journals Method for designing multi-input system identification signals using a compact time-frequency representation

2021 ◽  
Author(s):  
Mathias Stefan Roeser ◽  
Nicolas Fezans
Keyword(s):  
System Identification ◽  
Design Method ◽  
Flight Test ◽  
Compact Representation ◽  
Time Frequency ◽  
Stability And Control ◽  
Frequency Representation ◽  
Aerodynamic Parameters ◽  
Definition Of ◽  
And Control

AbstractA flight test campaign for system identification is a costly and time-consuming task. Models derived from wind tunnel experiments and CFD calculations must be validated and/or updated with flight data to match the real aircraft stability and control characteristics. Classical maneuvers for system identification are mostly one-surface-at-a-time inputs and need to be performed several times at each flight condition. Various methods for defining very rich multi-axis maneuvers, for instance based on multisine/sum of sines signals, already exist. A new design method based on the wavelet transform allowing the definition of multi-axis inputs in the time-frequency domain has been developed. The compact representation chosen allows the user to define fairly complex maneuvers with very few parameters. This method is demonstrated using simulated flight test data from a high-quality Airbus A320 dynamic model. System identification is then performed with this data, and the results show that aerodynamic parameters can still be accurately estimated from these fairly simple multi-axis maneuvers.

scholarly journals Improving Human Activity Recognition Performance by Data Fusion and Feature Engineering

Sensors ◽  
2021 ◽  
Vol 21 (3) ◽  
pp. 692
Author(s):  
Jingcheng Chen ◽  
Yining Sun ◽  
Shaoming Sun
Keyword(s):  
Feature Selection ◽  
Activity Recognition ◽  
Human Activity ◽  
Joint Angle ◽  
Recognition Performance ◽  
Wearable Sensors ◽  
Estimation Algorithm ◽  
Recognition System ◽  
Human Activity Recognition ◽  
Time Frequency

Human activity recognition (HAR) is essential in many health-related fields. A variety of technologies based on different sensors have been developed for HAR. Among them, fusion from heterogeneous wearable sensors has been developed as it is portable, non-interventional and accurate for HAR. To be applied in real-time use with limited resources, the activity recognition system must be compact and reliable. This requirement can be achieved by feature selection (FS). By eliminating irrelevant and redundant features, the system burden is reduced with good classification performance (CP). This manuscript proposes a two-stage genetic algorithm-based feature selection algorithm with a fixed activation number (GFSFAN), which is implemented on the datasets with a variety of time, frequency and time-frequency domain features extracted from the collected raw time series of nine activities of daily living (ADL). Six classifiers are used to evaluate the effects of selected feature subsets from different FS algorithms on HAR performance. The results indicate that GFSFAN can achieve good CP with a small size. A sensor-to-segment coordinate calibration algorithm and lower-limb joint angle estimation algorithm are introduced. Experiments on the effect of the calibration and the introduction of joint angle on HAR shows that both of them can improve the CP.

Sign in / Sign up

Export Citation Format

Share Document