Fault Diagnosis of Electric Two-Wheeler Under Pragmatic Operating Conditions Using Wavelet Synchrosqueezing Transform and CNN

Choudhary, Anurag; Mian, Tauheed; Fatima, Shahab; Panigrahi, Bijaya Ketan

doi:10.1109/jsen.2023.3239383

Cited by 13 publications

(2 citation statements)

References 49 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…identifying and diagnosing faults in IMs before they lead to motor failure, and include techniques such as vibration analysis, Motor Current Signal Analysis (MCSA), and motor current signature analysis [8,[10][11][12][13][14][15][16][17]. By implementing these fault diagnosis methods, the reliability of IMs can be enhanced, thereby improving the overall performance and safety of EVs.…”

Section: Of 17mentioning

confidence: 99%

Mathematical Decomposition of Fault Signature using Multinomial Identity and Differentiating for Detection of Broken Rotor Bar in Induction Motor

Behnamnia¹,

Mardaneh²,

Jamshidpour³

2023

Preprint

View full text Add to dashboard Cite

Fault detection at the early stage of development is of great importance in the maintenance of electric motors. In this regard, Motor Current Signature Analysis (MCSA) is cited in many articles to detect rotor bar fracture. The popularity of MCSA is due to its non-invasive and non-destructive nature, simplicity, and, compatibility with several signal processing tools. The vast variety of the signal processing tools which are commonly used for detecting the broken rotor bar fault is based on the Fast Fourier Transform (FFT) of the stator current signal and analysis of the produced Fourier spectrum; however, because of the small amplitude at the fault frequency relative to the main frequency amplitude, the former cannot be easily discriminated from the latter. Therefore, the peak at the fault frequency is hidden in the shadow of the main frequency component, hence the fault will not be detectable. As a solution to vanish the base frequency component, a method based on the Algebraic Identity of trinomial expansion is proposed in this paper which enables us to display the difference between the frequency of the fault signature and the base frequency. In this article, the fault of the broken rotor bars of the squirrel cage induction motor is revealed by the proposed method. In addition, a frequency weighting technique is presented to magnify the component at the fault related frequency compared to the lower frequencies. To validate the proposed methods, they are examined on the laboratory data obtained from three different operating conditions including the direct online start, the direct torque control, and the scalar control and the results show the ability of the proposed methods in fault detection of an induction motor.

show abstract

Section: Of 17mentioning

confidence: 99%

Mathematical Decomposition of Fault Signature using Multinomial Identity and Differentiating for Detection of Broken Rotor Bar in Induction Motor

Behnamnia¹,

Mardaneh²,

Jamshidpour³

2023

Preprint

View full text Add to dashboard Cite

show abstract

“…Therefore, most studies combine WT with other methods. To identify various bearing defects of electric vehicle engines, Choudhary et al [ 4 ] used the wavelet synchrosqueezing transform method to decompose vibration signals and convert them into time–frequency representations. Subsequently, they used a convolutional neural network for fault diagnosis.…”

Section: Introductionmentioning

confidence: 99%

Fault Diagnosis Method for Rolling Bearings Based on Grey Relation Degree

Mao,

Xin,

Zang

et al. 2024

Entropy

View full text Add to dashboard Cite

Aiming at the difficult problem of extracting fault characteristics and the low accuracy of fault diagnosis throughout the full life cycle of rolling bearings, a fault diagnosis method for rolling bearings based on grey relation degree is proposed in this paper. Firstly, the subtraction-average-based optimizer is used to optimize the parameters of the variational mode decomposition algorithm. Secondly, the vibration signals of bearings are decomposed by using the optimized results, and the feature vector of the intrinsic mode function component corresponding to the minimum envelope entropy is extracted. Finally, the grey proximity and similarity relation degree based on standard distance entropy are weighted to calculate the grey comprehensive relation degree between the feature vector of vibration signals and each standard state. By comparing the results, the diagnosis of different fault states and degrees of rolling bearings is realized. The XJTU-SY dataset was used for experimentation, and the results show that the proposed method achieves a diagnostic accuracy of 95.24% and has better diagnosis performance compared to various algorithms. It provides a reference for the fault diagnosis of rolling bearings throughout the full life cycle.

show abstract