scholarly journals Fault Diagnosis of a Reciprocating Compressor Air Valve Based on Deep Learning

2020 ◽  
Vol 10 (18) ◽  
pp. 6596
Author(s):  
Shungen Xiao ◽  
Ang Nie ◽  
Zexiong Zhang ◽  
Shulin Liu ◽  
Mengmeng Song ◽  
...  
Keyword(s):  
Machine Learning ◽  
Fault Diagnosis ◽  
Complex Systems ◽  
Single Channel ◽  
Vibration Signal ◽  
Extraction Methods ◽  
Reciprocating Compressor ◽  
Deep Convolutional Neural Networks ◽  
Phase Signal ◽  
Diagnosis Accuracy

With the development of machine learning in recent years, the application of machine learning to machine fault diagnosis has become increasingly popular. Applying traditional feature extraction methods for complex systems will weaken the characterization capacity of features, which are not conducive to subsequent classification work. A reciprocating compressor is a complex system. In order to improve the fault diagnosis accuracy of complex systems, this paper does not use traditional fault diagnosis methods and applies deep convolutional neural networks (CNNs) to process this nonlinear and non-stationary fault signal. The valve fault data is obtained from the reciprocating compressor test bench of the Daqing Natural Gas Company. Firstly, the single-channel vibration signal is collected on the reciprocating compressor and the one-dimensional CNN (1-D CNN) is used for fault diagnosis and compared with the traditional model to verify the effectiveness of the 1-D CNN. Next, the collected eight channels signals (three channels of vibration signals, four channels of pressure signals, one channel key phase signal) are applied by 1-D CNN and 2-D CNN for fault diagnosis to verify the CNN that it is still suitable for multi-channel signal processing. Finally, further study on the influence of the input of different channel signal combinations on the model diagnosis accuracy is carried out. Experiments show that the seven-channel signal (three-channel vibration signal, four-channel pressure signal) with the key phase signal removed has the highest diagnostic accuracy in the 2-D CNN. Therefore, proper deletion of useless channels can not only speed up network operations but also improve diagnosis accuracy.

Fault Diagnosis with Mult-Signal Test for Reciprocating Compressor

2015 ◽  
Vol 741 ◽  
pp. 294-297
Author(s):  
Jiang Ping Wang ◽  
Teng Fei Duan
Keyword(s):  
Fault Diagnosis ◽  
Ultrasonic Wave ◽  
High Frequency ◽  
Vibration Signal ◽  
Pressure Curve ◽  
Reciprocating Compressor ◽  
High Frequency Vibration ◽  
Low Pass ◽  
Diagnosis Method ◽  
Band Pass

High frequency vibration signal, cylinder-inside pressure and ultrasonic wave are monitored under working condition of a reciprocating compressor. High frequency vibration signal and ultrasonic wave are processed using high frequency band-pass filtering, and then enveloped. The pressure signal is filtered with a low-pass digital filter. The fault causes of the machine can be determined by the enveloped waveform of vibration signal and ultrasonic wave, as well as the pressure curve comprehensively. We can confirm the truth that this fault diagnosis method is accurate and reliable for reciprocating compressors through a number of the results of fault diagnosis.

scholarly journals Motor Fault Diagnosis Algorithm Based on Wavelet and Attention Mechanism

2021 ◽  
Vol 2021 ◽  
pp. 1-9
Author(s):  
Yong Yan ◽  
Qiang Liu ◽  
Xiao qin Gao
Keyword(s):  
Fault Diagnosis ◽  
Frequency Band ◽  
Signal To Noise Ratio ◽  
Vibration Signal ◽  
Attention Mechanism ◽  
Time Dimension ◽  
High Frequency Signal ◽  
Diagnosis Accuracy ◽  
Diagnosis Algorithm ◽  
Motor Vibration

In order to improve the maintenance efficiency of the motor and realize the real-time fault diagnosis function of the motor, a motor fault diagnosis algorithm based on wavelet and attention mechanism is proposed. Firstly, the motor vibration signal is decomposed by wavelet transform, and the high-frequency signal is denoised to improve the signal-to-noise ratio. Secondly, the frequency band and time dimension after wavelet decomposition are taken as input data, the convolution neural network is used to fuse the frequency band features of data, and the bidirectional gated loop unit is used to fuse the time series features. Then, the attention mechanism is used to adaptively integrate the features of different time points. Finally, motor fault diagnosis and prediction are realized by classifier recognition. Experimental results show that, compared with the existing deep learning fault diagnosis model, this method has higher diagnosis accuracy and can accurately diagnose the running state of the motor.

scholarly journals Rolling Element Bearing Fault Detection using Statistical Features and Ensemble Classifiers

2020 ◽  
Vol 9 (3) ◽  
pp. 350-358
Keyword(s):  
Machine Learning ◽  
Fault Diagnosis ◽  
Vibration Signal ◽  
Rolling Element Bearing ◽  
Machine Learning Techniques ◽  
Ensemble Classifiers ◽  
Statistical Features ◽  
Bearing Fault ◽  
Rolling Element ◽  
Element Bearing

Rolling element bearing health condition is monitored by analysing its vibration signature. Raw vibration signal picked up through suitably placed accelerometers is difficult to analyse hence many signal processing techniques have been proposed and developed by researchers to process the data for suitably extracting an effective signal feature set. Various machine learning techniques have been used for interpretation and accurate fault diagnosis using this extracted feature set. In this study “Empirical mode decomposition” is used for pre-processing the raw vibration data. Six “Statistical features” are extracted from the best Intrinsic mode function obtained through EMD and “Ensemble machine learning classifiers” are used for bearing fault diagnosis. A stacked ensemble of five classifiers is proposed for accurate fault diagnosis and results are compared with conventional ensemble classifiers to prove its effectiveness

scholarly journals Combining Multi-Scale Wavelet Entropy and Kernelized Classification for Bearing Multi-Fault Diagnosis

Entropy ◽  
2019 ◽  
Vol 21 (2) ◽  
pp. 152 ◽  
Author(s):  
Nibaldo Rodriguez ◽  
Pablo Alvarez ◽  
Lida Barba ◽  
Guillermo Cabrera-Guerrero
Keyword(s):  
Feature Extraction ◽  
Wavelet Packet ◽  
Vibration Signal ◽  
Work Conditions ◽  
Extraction Methods ◽  
Rolling Element Bearing ◽  
Bearing Failure ◽  
Wavelet Entropy ◽  
Multi Scale ◽  
Diagnosis Accuracy

Discriminative feature extraction and rolling element bearing failure diagnostics are very important to ensure the reliability of rotating machines. Therefore, in this paper, we propose multi-scale wavelet Shannon entropy as a discriminative fault feature to improve the diagnosis accuracy of bearing fault under variable work conditions. To compute the multi-scale wavelet entropy, we consider integrating stationary wavelet packet transform with both dispersion (SWPDE) and permutation (SWPPE) entropies. The multi-scale entropy features extracted by our proposed methods are then passed on to the kernel extreme learning machine (KELM) classifier to diagnose bearing failure types with different severities. In the end, both the SWPDE–KELM and the SWPPE–KELM methods are evaluated on two bearing vibration signal databases. We compare these two feature extraction methods to a recently proposed method called stationary wavelet packet singular value entropy (SWPSVE). Based on our results, we can say that the diagnosis accuracy obtained by the SWPDE–KELM method is slightly better than the SWPPE–KELM method and they both significantly outperform the SWPSVE–KELM method.

scholarly journals A Refined Composite Multivariate Multiscale Fluctuation Dispersion Entropy and Its Application to Multivariate Signal of Rotating Machinery

Entropy ◽  
2021 ◽  
Vol 23 (1) ◽  
pp. 128
Author(s):  
Chenbo Xi ◽  
Guangyou Yang ◽  
Lang Liu ◽  
Hongyuan Jiang ◽  
Xuehai Chen
Keyword(s):  
Fault Diagnosis ◽  
Single Channel ◽  
Vibration Signal ◽  
Rotating Machinery ◽  
Support Vector ◽  
Data Set ◽  
Rotary Machine ◽  
Gear Fault ◽  
Diagnosis Method ◽  
Low Sensitivity

In the fault monitoring of rotating machinery, the vibration signal of the bearing and gear in a complex operating environment has poor stationarity and high noise. How to accurately and efficiently identify various fault categories is a major challenge in rotary fault diagnosis. Most of the existing methods only analyze the single channel vibration signal and do not comprehensively consider the multi-channel vibration signal. Therefore, this paper presents Refined Composite Multivariate Multiscale Fluctuation Dispersion Entropy (RCMMFDE), a method which extracts the recognition information of multi-channel signals with different scale factors, and the refined composite analysis ensures the recognition stability. The simulation results show that this method has the characteristics of low sensitivity to signal length and strong anti-noise ability. At the same time, combined with Joint Mutual Information Maximisation (JMIM) and support vector machine (SVM), RCMMFDE-JMIM-SVM fault diagnosis method has been proposed. This method uses RCMMFDE to extract the state characteristics of the multiple vibration signals of the rotary machine, and then uses the JMIM method to extract the sensitive characteristics. Finally, different states of the rotary machine are classified by SVM. The validity of the method is verified by the composite gear fault data set and bearing fault data set. The diagnostic accuracy of the method is 99.25% and 100.00%. The experimental results show that RCMMFDE-JMIM-SVM can effectively recognize multiple signals.

scholarly journals An adaptive approach to machine learning for compact particle accelerators

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Alexander Scheinker ◽  
Frederick Cropp ◽  
Sergio Paiagua ◽  
Daniele Filippetto
Keyword(s):  
Machine Learning ◽  
Complex Systems ◽  
National Laboratory ◽  
Training Data ◽  
Particle Accelerators ◽  
Time Varying ◽  
Input Beam ◽  
Output Beam ◽  
Deep Convolutional Neural Networks ◽  
Time Varying Systems

AbstractMachine learning (ML) tools are able to learn relationships between the inputs and outputs of large complex systems directly from data. However, for time-varying systems, the predictive capabilities of ML tools degrade if the systems are no longer accurately represented by the data with which the ML models were trained. For complex systems, re-training is only possible if the changes are slow relative to the rate at which large numbers of new input-output training data can be non-invasively recorded. In this work, we present an approach to deep learning for time-varying systems that does not require re-training, but uses instead an adaptive feedback in the architecture of deep convolutional neural networks (CNN). The feedback is based only on available system output measurements and is applied in the encoded low-dimensional dense layers of the encoder-decoder CNNs. First, we develop an inverse model of a complex accelerator system to map output beam measurements to input beam distributions, while both the accelerator components and the unknown input beam distribution vary rapidly with time. We then demonstrate our method on experimental measurements of the input and output beam distributions of the HiRES ultra-fast electron diffraction (UED) beam line at Lawrence Berkeley National Laboratory, and showcase its ability for automatic tracking of the time varying photocathode quantum efficiency map. Our method can be successfully used to aid both physics and ML-based surrogate online models to provide non-invasive beam diagnostics.

scholarly journals A Parameter-Optimized DBN Using GOA and Its Application in Fault Diagnosis of Gearbox

2020 ◽  
Vol 2020 ◽  
pp. 1-11 ◽  
Author(s):  
Jingbo Gai ◽  
Junxian Shen ◽  
He Wang ◽  
Yifan Hu
Keyword(s):  
Feature Extraction ◽  
Fault Diagnosis ◽  
Optimal Parameter ◽  
Fitness Function ◽  
Great Influence ◽  
Vibration Signal ◽  
Extraction Methods ◽  
Grasshopper Optimization Algorithm ◽  
Adaptive Feature Extraction ◽  
Self Adaptive

Aiming at the problems of poor self-adaptive ability in traditional feature extraction methods and weak generalization ability in single classifier under big data, an internal parameter-optimized Deep Belief Network (DBN) method based on grasshopper optimization algorithm (GOA) is proposed. First, the minimum Root Mean Square Error (RMSE) in the network training is taken as the fitness function, in which GOA is used to search for the optimal parameter combination of DBN. After that the learning rate and the number of batch learning in DBN which have great influence on the training error would be properly selected. At the same time, the optimal structure distribution of DBN is given through comparison. Then, FFT and linear normalization are introduced to process the original vibration signal of the gearbox, preprocess the data from multiple sensors and construct the input samples for DBN. Finally, combining with deep learning featured by powerful self-adaptive feature extraction and nonlinear mapping capabilities, the obtained samples are input into DBN for training, and the fault diagnosis model for gearbox based on DBN would be established. After several tests with the remaining samples, the diagnosis rate of the model could reach over 99.5%, which is far better than the traditional fault diagnosis method based on feature extraction and pattern recognition. The experimental results show that this method could effectively improve the self-adaptive feature extraction ability of the model as well as its accuracy of fault diagnosis, which has better generalization performance.

scholarly journals A Novel Fault Diagnosis Method for Diesel Engine Based on MVMD and Band Energy

2020 ◽  
Vol 2020 ◽  
pp. 1-17
Author(s):  
Cheng Gu ◽  
Xinyong Qiao ◽  
Ying Jin ◽  
Yanbin Liu
Keyword(s):  
Fault Diagnosis ◽  
Diesel Engine ◽  
Working Conditions ◽  
Single Channel ◽  
Recognition Rate ◽  
Great Influence ◽  
Vibration Signal ◽  
Measuring Point ◽  
Band Energy ◽  
Vibration Signal Analysis

Vibration signal, as an important means for diesel engine condition detection and fault diagnosis, has attracted attention for many years. In traditional vibration signal analysis, most processing methods are for single-channel data. However, single-channel vibration signal cannot reflect the operating information of the diesel engine comprehensively because diesel engine vibration is coupled by multiple source signals. This paper proposes the MVMD band energy method for fault diagnosis by four channels of vibration signals. First, the original multivariate signals are decomposed adaptively by MVMD, which obtains a series of components with modal alignment. Then, the band energy values of each measuring point are calculated as the fault characteristics. Finally, SVM is used to realize the diagnosis and identification of diesel engine misfire. The working conditions have a great influence on the vibration signal of the cylinder. In order to obtain the best diagnostic working conditions, six working conditions are set for testing. The result shows that the fault identification rate is highest under the 1500 rpm and 50% load working condition. The fault recognition rate of this method reaches more than 99%, which is superior to the other four common methods.

Simultaneous Fault Diagnosis of Main Retarder Using Improved Paired Relevance Vector Machine Based on Multi-Kernel Learning

2014 ◽  
Vol 614 ◽  
pp. 339-344
Author(s):  
Qing Ye ◽  
Hao Pan
Keyword(s):  
Fault Diagnosis ◽  
Cross Validation ◽  
Computational Cost ◽  
Vibration Signal ◽  
Relevance Vector Machine ◽  
Kernel Learning ◽  
Diagnosis Method ◽  
Diagnosis Accuracy ◽  
Fault Diagnosis System ◽  
Low Computational Cost

According to the practical requirement of auto manufacturer, excellent fault diagnosis system aiming at simultaneous fault is indispensable for main retarder of automobile. This paper proposes a novel diagnosis method which employs wavelet package transform and sample entropy to achieve feature extraction, later utilize relevance vector machine to construct a set of paired classifiers. Considering that features extracted from vibration signal are multiple and heterogeneous, we combine multi-kernel learning and relevance vector machine together and optimize kernel function parameters by using incremental learning, cross validation and genetic algorithm. Comparing with SVM and PNN, the experiment results verify high diagnosis accuracy and low computational cost of the proposed method.

Sign in / Sign up

Export Citation Format

Share Document