scholarly journals Fault Diagnosis of Reciprocating Compressor Based on Convolutional Neural Networks with Multisource Raw Vibration Signals

2019 ◽  
Vol 2019 ◽  
pp. 1-7 ◽  
Author(s):  
Hong-bai Yang ◽  
Jiang-an Zhang ◽  
Lei-lei Chen ◽  
Hong-li Zhang ◽  
Shu-lin Liu
Keyword(s):  
Neural Network ◽  
Fault Diagnosis ◽  
Wavelet Packet ◽  
Recognition Rate ◽  
Back Propagation ◽  
Petroleum Industry ◽  
Back Propagation Neural Network ◽  
Reciprocating Compressor ◽  
Vibration Signals ◽  
Fault Recognition

Reciprocating compressors are widely used in petroleum industry. Due to containing complex nonlinear signal, it is difficult to extract the fault features from its vibration signals. This paper proposes a new method named Convolutional Neural Network based on Multisource Raw vibration signals (MSRCNN). The proposed method uses multisource raw vibration signals collected by several sensors as input and uses the designed CNN to operate both the feature extraction and classification. The gas valve signals of reciprocating compressor in different states are used as the experimental data. In order to test the effectiveness of the proposed method, it is compared with the traditional BP (Back-Propagation) neural network fault diagnosis method based on power spectrum energy and wavelet packet energy. In order to further test the antinoise performance of the proposed method, some noisy signals with different signal-to-noise ratios were constructed by adding white noise into sampled signals for testing. The results show that the MSRCNN model has higher fault recognition rate than the traditional methods. This indicates that the MSRCNN method not only has good fault recognition effect, but also has certain antinoise performance.

scholarly journals Applying Energy-Equal Entropy of Wavelet Packet to Diagnose Circuit Breaker Faults

2014 ◽  
Vol 8 (1) ◽  
pp. 445-452
Author(s):  
Liu Mingliang ◽  
Wang Keqi ◽  
Sun Laijun ◽  
Zhang Jianfeng
Keyword(s):  
Neural Network ◽  
Fault Diagnosis ◽  
Bp Neural Network ◽  
Wavelet Packet ◽  
Circuit Breaker ◽  
Back Propagation ◽  
Vibration Signal ◽  
Characteristic Parameter ◽  
Vibration Signals

Aiming to better reflect features of machinery vibration signals of high-voltage (HV) circuit breaker (CB), a new method is proposed on the basis of energy-equal entropy of wavelet packet(WP). First of all, three-layer wavelet packet decomposes vibration signal, reconstructing 8 nodes of signals in the 3rd layer. Then, the vector is extracted with energy-equal entropy of reconstructed signals. At last, the simple back-propagation (BP) neural network for fault diagnosis contributes to classification of the characteristic parameter. This technology is the basis of a number of patents and patents pending, which is experimentally demonstrated by the significant improvement of diagnose faults.

Power Transform Fault Diagnosis Based on the Adaptive Mutation Particle Swarm Optimization

2013 ◽  
Vol 756-759 ◽  
pp. 3804-3808
Author(s):  
Zhi Mei Duan ◽  
Jia Tang Cheng
Keyword(s):  
Neural Network ◽  
Particle Swarm Optimization ◽  
Fault Diagnosis ◽  
Power Transformer ◽  
Recognition Rate ◽  
Particle Swarm ◽  
Adaptive Mutation ◽  
Bp Network ◽  
Swarm Optimization ◽  
Fault Recognition

In order to improve the accuracy of fault diagnosis of power transformer, in this paper, a method is proposed that optimize the weight of BP neural network by adaptive mutation particle swarm optimization (AMPSO). According to the characteristic of transformer fault, the optimized neural network is used to diagnose fault of the power transformer. Individual particles action is amended by this algorithm and local minima problems of the standard PSO and BP network are overcooked. The experimental results show that, the method can classify transformer faults, and effectively improve the fault recognition rate.

Power Quality Disturbance Classification Using Wavelet Packet Transform and Back Propagation Algorithm

2016 ◽  
Vol 13 (10) ◽  
pp. 7099-7109
Author(s):  
M. K Elango ◽  
A Jagadeesan ◽  
K. Mohana Sundaram
Keyword(s):  
Neural Network ◽  
Power Quality ◽  
Virtual Instrument ◽  
Wavelet Packet ◽  
Back Propagation ◽  
Back Propagation Neural Network ◽  
Wavelet Packet Transform ◽  
Back Propagation Algorithm ◽  
Propagation Algorithm ◽  
Power Quality Disturbances

This paper develops a real time solution for detecting the Power Quality events. Fourteen events are generated through experimental setup and the signals are acquired through a voltage Data Acquisition Card, NI DAQ-9225, controlled by a Virtual Instrument software package. The features extracted from the Wavelet Transformation are fed into the Back Propagation Neural Network for training. By the virtue of a Neural Network property, it gets self-adapted and self-learned aiding in automatic classification of Power Quality Events. A combination of Wavelet Transform technique and Neural Networks are employed to detect and characterize the Power Quality Disturbances. The result obtained shows the effectiveness of the Wavelet Packet Transform based Back Propagation algorithm in classifying the Power Quality Disturbances. The results produced by the proposed methodology based Back Propagation Algorithm is verified with the Power Quality Analyser.

scholarly journals Fault Diagnosis of Reciprocating Compressor Valve Based on Transfer Learning Convolutional Neural Network

2021 ◽  
Vol 2021 ◽  
pp. 1-13
Author(s):  
Fu-Yan Guo ◽  
Yan-Chao Zhang ◽  
Yue Wang ◽  
Pei-Jun Ren ◽  
Ping Wang
Keyword(s):  
Neural Network ◽  
Fault Diagnosis ◽  
Convolutional Neural Network ◽  
Transfer Learning ◽  
Recognition Rate ◽  
Vital Role ◽  
Stable Operation ◽  
Reciprocating Compressor ◽  
Gas Temperature ◽  
Label Prediction

Reciprocating compressors play a vital role in oil, natural gas, and general industrial processes. Their safe and stable operation directly affects the healthy development of the enterprise economy. Since the valve failure accounts for 60% of the total failures when the reciprocating compressor fails, it is of great significance to quickly find and diagnose the failure type of the valve for the fault diagnosis of the reciprocating compressor. At present, reciprocating compressor valve fault diagnosis based on deep neural networks requires sufficient labeled data for training, but valve in real-case reciprocating compressor (VRRC) does not have enough labeled data to train a reliable model. Fortunately, the data of valve in laboratory reciprocating compressor (VLRC) contains relevant fault diagnosis knowledge. Therefore, inspired by the idea of transfer learning, a fault diagnosis method for reciprocating compressor valves based on transfer learning convolutional neural network (TCNN) is proposed. This method uses convolutional neural network (CNN) to extract the transferable features of gas temperature and pressure data from VLRC and VRRC and establish pseudolabels for VRRC unlabeled data. Three regularization terms, the maximum mean discrepancy (MMD) of the transferable features of VLRC and VRRC data, the error between the VLRC sample label prediction and the actual label, and the error between the VRRC sample label prediction and the pseudolabel, are proposed. Their weighted sum is used as an objective function to train the model, thereby reducing the distribution difference of domain feature transfer and increasing the distance between learning feature classes. Experimental results show that this method uses VLRC data to identify the health status of VRRC, and the fault recognition rate can reach 98.32%. Compared with existing methods, this method has higher diagnostic accuracy, which proves the effectiveness of this method.

FFBPNN-BASED HIGH DROWSINESS CLASSIFICATION USING EMG AND WPT

2020 ◽  
Vol 32 (03) ◽  
pp. 2050023 ◽  
Author(s):  
Mousa Kadhim Wali
Keyword(s):  
Neural Network ◽  
Wavelet Packet ◽  
Back Propagation ◽  
Back Propagation Neural Network ◽  
Road Accidents ◽  
Mean Values ◽  
Physiological Signal ◽  
Feed Forward Back Propagation ◽  
Emg Signal ◽  
Amplitude Level

The detection of drowsiness level is important because it is the main reason for fatal road accidents. Electromyography of the upper arm and shoulder is an important physiological signal affected by drivers’ drowsiness, in which its amplitude level and frequency band of the sleep-deprived case are different than those of the alert state. Therefore depending on electromyography (EMG), its drowsiness frequency (80–100[Formula: see text]Hz) was detected in order to determine high drowsiness state based on wavelet packet transform (WPT) which decomposes the EMG signal into its approximation and detail coefficients up to level 4 using db2, db7, sym5 and coif5 wavelets. In this research after extraction, the two higher order statistical features, kurtosis and skewness, are computed from 3[Formula: see text]s window of the three EMG channels, and analysis of variance test is used to check whether their mean values are different for the different classes as both [Formula: see text]-values are less than 0.005 under db2 wavelet. Therefore, they were supplied to feed forward back propagation neural network (FFBPNN) as this type of neural network is used for distinguishing and classification purposes for different objects. They obtained an accuracy of 75% for detecting high levels among other levels of normal and low drowsiness with an average sensitivity of 78.63% and specificity of 75.97% because the spectrum of the EMG alert (non-drowsiness) signal of 80–100 Hz is different from that of drowsy 80–90[Formula: see text]Hz and high drowsy 78–95[Formula: see text]Hz signals.

Application of Time-Frequency Analysis and Back-Propagation Neural Network in the Lung Sound Signal Recognition

2012 ◽  
Vol 190-191 ◽  
pp. 927-930
Author(s):  
Mei Yung Chen ◽  
Chien Chou Huang
Keyword(s):  
Neural Network ◽  
Frequency Analysis ◽  
Recognition Rate ◽  
Back Propagation ◽  
Back Propagation Neural Network ◽  
Visual Signal ◽  
Lung Sound ◽  
Time Frequency Analysis ◽  
Time Frequency ◽  
Breath Sounds

In the diagnosis of the respiratory diseases, auscultation is a non-invasive and convenient diagnostic method. In the digital auscultation analysis, what method we use to analyze the lung signals which microphone recorded will affect the results of the experiment greatly. The purpose of this study is to use frequency analysis and time-frequency analysis to analyze the six lung sound signals, which are vesicular breath sounds, bronchial breath sounds, crackle, and wheeze. Finally, the study transformed the analysis results into the characteristic images, and put them to the back propagation neural network for training. After that, the study compares the results of the two methods. We also analyze the realistic lung sound signals and simulated lung sound signals, and compare the results finally. First, we use the piezoelectric microphone and data acquisition card NI-PXI 4472B to acquire LS signals, and signals preprocessing. Then we use Visual Signal to analyze the lung sound signals by time-frequency analysis. We also analyze the lung sound signals which are from the auscultation teaching website. Finally we compare the result of two kinds of signals, and assess their similarity and accuracy by the test of back-propagation neural network. According to the result of this study, we found that time-frequency analysis provide much information about the lung signals, and are more suitable as a basis of diagnosis, and increase the recognition rate of the back-propagation neural network.

Sign in / Sign up

Export Citation Format

Share Document