scholarly journals A Novel Arc Fault Detection Method Integrated Random Forest, Improved Multi-scale Permutation Entropy and Wavelet Packet Transform

Electronics ◽  
2019 ◽  
Vol 8 (4) ◽  
pp. 396 ◽  
Author(s):  
Zhendong Yin ◽  
Li Wang ◽  
Yaojia Zhang ◽  
Yang Gao
Keyword(s):  
Random Forest ◽  
Fault Detection ◽  
Detection Method ◽  
Wavelet Packet ◽  
Wavelet Packet Transform ◽  
Permutation Entropy ◽  
High Dimensional ◽  
Multi Scale ◽  
Fault Features ◽  
Load Types

Arc faults are one of the important causes of electric fires. In order to solve the problem of randomness, diversity, the concealment of series arc faults and to improve the detection accuracy, a novel arc fault detection method integrated random forest (RF), improved multi-scale permutation entropy (IMPE) and wavelet packet transform (WPT) are designed. Firstly, singular value decomposition (SVD) was applied to filter the current signal and then the high-dimensional fault features were constructed by extracting IMPE, the wavelet packet energy and the wavelet packet energy-entropy. Afterward, the high-dimensional fault features were employed to train the RF to realize the arc fault detection of different load types and the experimental results verify the effectiveness of the arc fault detection method designed in this paper. Finally, the comparative experiments demonstrates that the RF shows better performance in arc fault detection compared to the back-propagation neural network (BPNN) and least squares support vector machines (LSSVM), and that the experiments of transient events indicate that RF is able to effectively avoid incorrectly detecting different load types during the start operations and stop operations.

Looseness condition feature extraction of viscoelastic sandwich structure using dual-tree complex wavelet packet-based deep autoencoder network

2019 ◽  
Vol 19 (3) ◽  
pp. 873-884
Author(s):  
Yue Si ◽  
Zhousuo Zhang ◽  
Chuiqing Kong ◽  
Shujuan Li ◽  
Guigeng Yang ◽  
...  
Keyword(s):  
Feature Extraction ◽  
Sandwich Structure ◽  
Detection Method ◽  
Sandwich Structures ◽  
Wavelet Packet ◽  
Wavelet Packet Transform ◽  
Vibration Response ◽  
Permutation Entropy ◽  
Response Signal ◽  
Complex Wavelet

It is significant to perform looseness condition detection of viscoelastic sandwich structures to avoid serious accidents. Due to the multilayer characteristic of the viscoelastic sandwich structure, the vibration response signal of such structures is nonlinear and nonstationary. Furthermore, the looseness condition feature signal contained in the vibration response signal is very puny. Condition feature extraction has become a challenging task in the looseness condition detection of viscoelastic sandwich structures. Therefore, a novel method called dual-tree complex wavelet packet-based deep autoencoder network is proposed for this task. First, the vibration response signal of the viscoelastic sandwich structure is decomposed by dual-tree complex wavelet packet transform and the sub-band signals which contain rich energy are extracted. Then, the energies of the extracted sub-band signals are calculated to form a feature set. Finally, a deep autoencoder network is established to fuse the feature set, and the fused feature is viewed as the detection index to detect the looseness condition of the viscoelastic sandwich structure. The proposed method is applied to the connecting bolt looseness condition detection of the viscoelastic sandwich structure to validate its effectiveness. Compared with the detection method based on dual-tree complex wavelet packet transform and energy and the detection method based on dual-tree complex wavelet packet transform and permutation entropy, the results indicate that the effectiveness of the proposed method in this article is more superior to that of the other two methods.

Evaluation of a new proposal for an arcing fault detection method based on wavelet packet analysis

2004 ◽  
Vol 14 (3) ◽  
pp. 161-174 ◽  
Author(s):  
Andoni Lazkano ◽  
Jesus Ruiz ◽  
Elisabete Aramendi ◽  
Luis A. Leturiondo
Keyword(s):  
Fault Detection ◽  
Detection Method ◽  
Wavelet Packet ◽  
Wavelet Packet Analysis ◽  
Arcing Fault

Evaluation of Diagnostic Performance for the Gear Box Considering the Measurement Position Effect With Wavelet Packet Transform

2009 ◽  
Author(s):  
Young-Sun Hong ◽  
Gil-Yong Lee ◽  
Young-Man Cho ◽  
Sung-Hoon Ahn ◽  
Chul-Ki Song
Keyword(s):  
Wavelet Packet ◽  
Position Effect ◽  
Vibration Signal ◽  
Industrial Applications ◽  
Wavelet Packet Transform ◽  
Bevel Gear ◽  
Diagnostic Methods ◽  
Vibration Signals ◽  
Rotary Machinery ◽  
Fault Features

There has been much research into monitoring techniques for mechanical systems to ensure stable production levels in modern industries. This is particularly true for the diagnostic monitoring of rotary machinery, because faults in this type of equipment appear frequently and quickly cause severe problems. Such diagnostic methods are often based on the analysis of vibration signals because they are directly related to physical faults. Even though the magnitude of vibration signals depends on the measurement position, the effect of measurement position is generally not considered. This paper describes an investigation of the effect of the measurement position on the fault features in vibration signals. The signals for normal and broken bevel gears were measured at the base, gearbox, and bevel gear, simultaneously, of a machine fault simulator (MFS). These vibration signals were compared to each other and used to estimate the classification efficiency of a diagnostic method using wavelet packet transform. From this experiment, the fault features are more prominently in the vibration signal from the measurement position of the bevel gear than from the base and gearbox. The results of this analysis will assist in selecting the appropriate measurement position in real industrial applications and precision diagnostics.

Sign in / Sign up

Export Citation Format

Share Document