scholarly journals Unsupervised Detection of Anomalous Sound for Machine Condition Monitoring using Fully Connected U-Net

2021 ◽  
Vol 15 (1) ◽  
pp. 41-55
Author(s):  
Hoang Van Truong ◽  
Nguyen Chi Hieu ◽  
Pham Ngoc Giao ◽  
Nguyen Xuan Phong
Keyword(s):  
Reconstruction Error ◽  
Conventional Approach ◽  
Machine Condition ◽  
Machine Monitoring ◽  
Machine Condition Monitoring ◽  
Novel Approach ◽  
Baseline System ◽  
Optimized Model ◽  
Fully Connected ◽  
Spectrogram Feature

Anomaly detection in the sound from machines is an important task in machine monitoring. An autoencoder architecture based on the reconstruction error using a log-Mel spectrogram feature is a conventional approach for this domain. However, because of the non-stationary nature of some sounds from the target machine, such a conventional approach does not perform well in those circumstances. In this paper, we propose a novel approach regarding the choice of used features and a new auto-encoder architecture. We created the Mixed Feature, which is a mixture of different sound representations, and a new deep learning method called Fully-Connected U-Net, a form of autoencoder architecture. With experiments on the same dataset as the baseline system, using the same architecture for all types of machines, the experimental results showed that our methods outperformed the baseline system in terms of the AUC and pAUC evaluation metrics. The optimized model achieved 83.38% AUC and 64.51% pAUC on average overall machine types on the developed dataset and outperformed the published baseline by 13.43% AUC and 8.13% pAUC.

BASELINE-DIFFERENCING: A NOVEL APPROACH FOR BUILDING GENERALIZABLE OCEAN TURBINE RELIABILITY MODELS

2012 ◽  
Vol 19 (06) ◽  
pp. 1240005
Author(s):  
RANDALL WALD ◽  
TAGHI M. KHOSHGOFTAAR ◽  
BASSEM ALHALABI
Keyword(s):  
Clean Energy ◽  
Operating Conditions ◽  
Sensor Data ◽  
Machine Condition ◽  
Quality Models ◽  
Reliability Models ◽  
Machine Condition Monitoring ◽  
Novel Approach ◽  
The Difference ◽  
Machine Health

Ocean turbines are a promising new source of clean energy, but their remote and inhospitable environment (the open ocean) poses reliability challenges. Machine condition monitoring/prognostic health monitoring (MCM/PHM) systems assure the reliability of these turbines by detecting and predicting machine state. These MCM/PHM systems use sensor data (such as vibration information) to determine whether or not the machine is operating properly. However, not all sensor data corresponds to the machine state: some portions of the sensor signal are influenced by certain environmental conditions which do not directly relate to machine health. Therefore, models must be built which can detect system state regardless of these environmental operating conditions. The proposed baseline-differencing approach permits this by creating a baseline for different conditions (such that each baseline represents what the normal, healthy machine state looks like while in that operating condition) and using the difference of the observed data and this baseline to train and evaluate models. We present two case studies, both conducted on data from a dynamometer representing an ocean turbine, to demonstrate the improved predictive capabilities of models which incorporate baseline-differencing, compared to the models which use the nonbaselined data. The results show that significantly more high-quality models can be built with baseline-differencing.

Recognition on Rolling Bearing’s Condition of Grinding Machine by Using a Novel Approach-Wavelet Hilbert Marginal Spectrum

2007 ◽  
Vol 329 ◽  
pp. 773-778
Author(s):  
H.K. Li ◽  
X.J. Ma ◽  
Q.M. Ren ◽  
J.L. Zhao
Keyword(s):  
Pattern Recognition ◽  
Condition Monitoring ◽  
Early Stage ◽  
Rolling Bearing ◽  
Machine Condition ◽  
Nonlinear Characteristics ◽  
Machine Condition Monitoring ◽  
Novel Approach ◽  
Marginal Spectrum ◽  
Grinding Machine

Grinding machine condition monitoring is very important during the manufacturing process. Vibration analysis is usually used to its pattern recognition. But traditional signal analysis method limits the accuracy of recognition because of non-stationary and nonlinear characteristics. In this paper, a novel approach is presented in detail for grinding machine fault diagnosis. The method is based on the new developed Hilbert Marginal Spectrum and wavelet transform, named as wavelet-Hilbert marginal spectrum (WHMS). A rolling bearing’s pattern recognition of grinding machine is used to testify the effectiveness of this method, which can accurately detect flaw of the rolling bearing in early stage. Thus, it can be concluded that this promising method will contribute the development of grinding machine condition monitoring.

scholarly journals Benefits and Techniques for Machine Condition Monitoring in Industrial Environments

2010 ◽  
Author(s):  
Michael Denton
Keyword(s):  
Condition Monitoring ◽  
Business Case ◽  
Machine Condition ◽  
Maintenance System ◽  
Machine Monitoring ◽  
Machine Condition Monitoring ◽  
Wide Range ◽  
Signal Processing Algorithms ◽  
Plant Equipment ◽  
Processing Algorithms

Condition monitoring of plant machinery is becoming more common place. With new advanced signal processing algorithms and better machine life models proactive maintenance of citrus processing machinery allows avoiding unplanned downtime and catastrophic failure. It also avoids relying only on predictions and assuming the machine will break. This paper will discuss the main steps that are necessary in developing a plant machinery maintenance system and make a business case for implementing machine monitoring on a wide range of plant equipment. Paper published with permission.

Dynamic Processes of Rotating System with Coupling

2006 ◽  
Vol 113 ◽  
pp. 213-218 ◽  
Author(s):  
Vytautas Barzdaitis ◽  
Marijonas Bogdevicius ◽  
Rimantas Didžiokas
Keyword(s):  
Finite Element Method ◽  
Finite Element ◽  
Elastic Plate ◽  
Diagnostic System ◽  
Complex Model ◽  
Dynamic Processes ◽  
Machine Condition ◽  
Rotating System ◽  
Machine Condition Monitoring ◽  
Dynamics And Vibration

The article concerns dynamics and vibration of a high power rotating system with toothed wheel coupling that inserts elastic plate packets. Theoretical modeling and simulation of the rotating system with toothed wheel coupling have been provided by a finite element method. Complex model involved a rotating system, hydrodynamic bearings and coupling. The dynamics of semi-couplings and plates have been simulated. Experimental measurement of vibration has been measured with stationary machine condition monitoring and a diagnostic system.

scholarly journals An Illustration of New Methods in Machine Condition Monitoring, Part II: Adaptive outlier detection

2017 ◽  
Vol 842 ◽  
pp. 012059
Author(s):  
I. Antoniadou ◽  
K. Worden ◽  
S. Marchesiello ◽  
C. Mba ◽  
L. Garibaldi
Keyword(s):  
Outlier Detection ◽  
Condition Monitoring ◽  
Machine Condition ◽  
New Methods ◽  
Machine Condition Monitoring

scholarly journals Review of Electrical Machine Diagnostic Methods Applicability in the Perspective of Industry 4.0

2018 ◽  
Vol 14 (2) ◽  
pp. 108-116 ◽  
Author(s):  
Bilal Asad ◽  
Toomas Vaimann ◽  
Anton Rassõlkin ◽  
Ants Kallaste ◽  
Anouar Belahcen
Keyword(s):  
Industry 4.0 ◽  
Industrial Sector ◽  
Diagnostic Methods ◽  
Electrical Machine ◽  
Machine Condition ◽  
New Horizons ◽  
Machine Condition Monitoring ◽  
Cloud Environments ◽  
The Internet Of Things ◽  
Selection Of

AbstractDigitalization of the industrial sector and Industry 4.0 have opened new horizons in many technical fields, including electrical machine diagnostics and operation, as well as machine condition monitoring. This paper addresses a selection of electrical machine diagnostics methods that are applicable for the use in the perspective of Industry 4.0, to be used in hand with cloud environments and the possibilities granted by the Internet of Things. The need for further research and development in the field is pointed out. Some potentially applicable future approaches are presented.

Diagnosis of Machinery Fault Status using Transient Vibration Signal Parameters

2002 ◽  
Vol 8 (3) ◽  
pp. 321-335 ◽  
Author(s):  
Zhidong Chen ◽  
Chris K. Mechefske
Keyword(s):  
Vibration Signal ◽  
Rolling Element Bearings ◽  
Machine Condition ◽  
Transient Vibration ◽  
Vibration Signals ◽  
Model Based ◽  
Machine Condition Monitoring ◽  
Signal Parameters ◽  
Rolling Element ◽  
Efficient Machine

This paper reports the results of an investigation in which a Prony model based method is developed. The method shows potential for analysing transient vibration signals. An example is included that shows how the procedure was employed to analyse the transient vibration signals created from faulty low speed rolling element bearings. Spectral plots generated by applying the procedure to very short data samples, as well as trending parameters based on these spectral estimations and Prony parameters, are presented. An equation was also derived to quantitatively determine the fault status. It is shown that application of the Prony model based method has the potential to be an effective as well as efficient machine condition monitoring and diagnostic tool where short duration transient vibration signals are being generated.

Sign in / Sign up

Export Citation Format

Share Document