Diagnostic of Combined Mechanical and Electrical Faults in ASD-Powered Induction Motor Using MODWT and a Lightweight 1-D CNN

Jiménez-Guarneros, Magdiel; Morales-Perez, Carlos; Rangel-Magdaleno, José

doi:10.1109/tii.2021.3120975

Cited by 39 publications

(14 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…[ 121 ] presents a 1-D CNN-based approach to automatically learn features for rub-impact fault diagnosis from the raw vibration signals of a rotor system, and [ 114 ] establish a fault identification model based on the powerful feature extraction and complex data analysis abilities of 1D-CNN. Due to its inherent properties, many modern techniques for 2-D CNN can be imported into 1-D CNN for better signal feature extraction, such as attention [ 112 ], lightweight design [ 122 ], and dilated convolution [ 123 ].…”

Section: Part Ii: Supervised DL Methods For Intelligent Industrial Fdpmentioning

confidence: 99%

Deep Learning Techniques in Intelligent Fault Diagnosis and Prognosis for Industrial Systems: A Review

Qiu

Cui

Ping

et al. 2023

Sensors

View full text Add to dashboard Cite

Fault diagnosis and prognosis (FDP) tries to recognize and locate the faults from the captured sensory data, and also predict their failures in advance, which can greatly help to take appropriate actions for maintenance and avoid serious consequences in industrial systems. In recent years, deep learning methods are being widely introduced into FDP due to the powerful feature representation ability, and its rapid development is bringing new opportunities to the promotion of FDP. In order to facilitate the related research, we give a summary of recent advances in deep learning techniques for industrial FDP in this paper. Related concepts and formulations of FDP are firstly given. Seven commonly used deep learning architectures, especially the emerging generative adversarial network, transformer, and graph neural network, are reviewed. Finally, we give insights into the challenges in current applications of deep learning-based methods from four different aspects of imbalanced data, compound fault types, multimodal data fusion, and edge device implementation, and provide possible solutions, respectively. This paper tries to give a comprehensive guideline for further research into the problem of intelligent industrial FDP for the community.

show abstract

Section: Part Ii: Supervised DL Methods For Intelligent Industrial Fdpmentioning

confidence: 99%

Deep Learning Techniques in Intelligent Fault Diagnosis and Prognosis for Industrial Systems: A Review

Qiu

Cui

Ping

et al. 2023

Sensors

View full text Add to dashboard Cite

show abstract

“…Mechanism failure introduces periodic fluctuations in the load torque, and the corresponding oscillation component with the same frequency is induced in the motor current by the transmission path of closed-loop control and the electromagnetic interaction between the rotor and stator. [6] Jimenez-Guarneros et al [7] proposed a motor current-based diagnosis methodology involving maximal overlap DWT and a lightweight 1-D CNN, and tested its performance under different levels of load. Consequently, a high fault detection accuracy was obtained using this method.…”

Section: Introductionmentioning

confidence: 99%

Bearing fault diagnosis under non-stationary conditions based on a speed signal resonance component demodulation algorithm

Ye,

Zhang,

Yang

2023

Meas. Sci. Technol.

View full text Add to dashboard Cite

This paper proposes a novel bearing fault diagnosis indicator, the resonance component, for motor speed signals. The speed signal was modulated into a higher-frequency band using a double-inertia system, and the bearing fault information was carried and reserved in this high-frequency resonance component. Variational mode decomposition was then used to separate the components unrelated to the resonance based on the parameters optimized using the artificial bee colony algorithm. After envelope demodulation and angular resampling, the multipoint optimal minimum entropy deconvolution adjusted algorithm was utilized to enhance the amplitude of the order spectrum at the corresponding faulty order. Additionally, experiments were successfully conducted on both the outer and inner raceway bearing faults under time-varying conditions using the proposed method. The results indicate that the ordinary diagnosis method using the basic bearing fault characteristic frequency as an indicator is less effective.

show abstract

“…One of the most frequent faults are electrical defects in stator windings, which account for up to 40% of all cases [8]. A hybrid neural network (combination of self-organizing Kohonen (SOM) and multi-layer perceptron (MLP) networks) has been used to detect one or two faults but with no load [8], lightweight 1-D convolutional neural network (CNN) architecture along with maximal overlap discrete wavelet transform [9], [10] to detect electrical and/or mechanical defects in the rotor, the artificial neural network (ANN) aided by the Clarke transform when open-circuit fault detection in the IGBT inverter happens [11], neural network and hidden Markov model together (NN-HMM) to detect rotor eccentricity faults using current measurements [12], ANFIS-based neuro-fuzzy network also to detect mechanical defect in machine rotor [13], [14] application C4. 5 DT and MLP ANN has made it possible to determine short circuits under unbalanced grid and variable load [15], self-organizing maps neural networks based on active and reactive power components [16].…”

Section: Introductionmentioning

confidence: 99%

Neural network based detecting induction motor defects supplied by unbalanced grid

Dobroskok

Skakun

Belskii

et al. 2023

IJPEDS

View full text Add to dashboard Cite

<span lang="EN-US">Predictive diagnosis of motor defects can reduce repair and downtime costs of electrical equipment. This paper presents the results of a research on the effectiveness of the defect classification and recognition system based on convolutional neural network to detect defects in a squirrel-cage induction motor based on the stator supply voltage and output phase currents when supplied from an industrial grid of limited power with possible voltage asymmetry and harmonic distortion taken into account. In this work a simulation model, implemented in MATLAB/Simulink, is proposed to investigate unbalanced conditions. The mathematical model has been verified on a physical test bench and then used to create a database of currents from measured asymmetric grid voltages in the presence of defects such as short circuit between turns of one phase, rotor bar breakage, rotor eccentricity and bearing defects. The classification quality of all defects in the neural network was 72%, with the exclusion of intercoil short circuit defect and the merging of different bearing defects, the quality of the model was achieved 91%. A further increase in the quality of the defect recognition system is due to a building up modification of the neural network architecture.</span>

show abstract

Diagnostic of Combined Mechanical and Electrical Faults in ASD-Powered Induction Motor Using MODWT and a Lightweight 1-D CNN

Cited by 39 publications

References 30 publications

Deep Learning Techniques in Intelligent Fault Diagnosis and Prognosis for Industrial Systems: A Review

Deep Learning Techniques in Intelligent Fault Diagnosis and Prognosis for Industrial Systems: A Review

Bearing fault diagnosis under non-stationary conditions based on a speed signal resonance component demodulation algorithm

Neural network based detecting induction motor defects supplied by unbalanced grid

Contact Info

Product

Resources

About