An intelligent fault detection and classification technique based on variational mode decomposition-CNN for transmission lines installed with UPFC and wind farm

Biswas, Sauvik; Nayak, Paresh Kumar; Panigrahi, Bijaya Ketan; Pradhan, Gayadhar

doi:10.1016/j.epsr.2023.109526

Cited by 30 publications

(9 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The convolution layer employs weight sharing, significantly reducing the number of learning parameters and thereby mitigating the risk of overfitting [ 52 ]. Each convolutional layer in the network comprises multiple convolutional kernels, with the parameters of each kernel being optimized through the backpropagation algorithm [ 53 ]. Each convolution can be performed on the input sequence by convolving the equation as follows:

where x m is the input sequence, w ( k ) is the weight of the k th convolution kernel, and the size is L.…”

Section: Design Of Cable Discharge Classification Modelmentioning

confidence: 99%

Convolutional Neural Network-Based Pattern Recognition of Partial Discharge in High-Speed Electric-Multiple-Unit Cable Termination

Sun,

Wu,

Pan

et al. 2024

Sensors

View full text Add to dashboard Cite

Partial discharge detection is considered a crucial technique for evaluating insulation performance and identifying defect types in cable terminals of high-speed electric multiple units (EMUs). In this study, terminal samples exhibiting four typical defects were prepared from high-speed EMUs. A cable discharge testing system, utilizing high-frequency current sensing, was developed to collect discharge signals, and datasets corresponding to these defects were established. This study proposes the use of the convolutional neural network (CNN) for the classification of discharge signals associated with specific defects, comparing this method with two existing neural network (NN)-based classification models that employ the back-propagation NN and the radial basis function NN, respectively. The comparative results demonstrate that the CNN-based model excels in accurately identifying signals from various defect types in the cable terminals of high-speed EMUs, surpassing the two existing NN-based classification models.

show abstract

where x m is the input sequence, w ( k ) is the weight of the k th convolution kernel, and the size is L.…”

Section: Design Of Cable Discharge Classification Modelmentioning

confidence: 99%

Convolutional Neural Network-Based Pattern Recognition of Partial Discharge in High-Speed Electric-Multiple-Unit Cable Termination

Sun,

Wu,

Pan

et al. 2024

Sensors

View full text Add to dashboard Cite

show abstract

“…According to the projected voltages at the ends of M and N, the fault ranging function was constructed by combining equation (8) and solved by the adaptive golden section search method. The calculation results are presented in figure 10.…”

Section: Example Analysismentioning

confidence: 99%

“…Due to the typical location of wind farms being in remote areas with long lines and complex environments, precise fault location is essential for ensuring stable operation of the power system. The main fault location methods in transmission lines currently include the impedance method [6,7], traveling wave method [8,9], and fault analysis method [10,11]. However, limited research has been conducted on fault location for wind power transmission lines.…”

Section: Introductionmentioning

confidence: 99%

Fault distance measurement method for wind power transmission lines based on improved NSGA II

Qi,

Zhang,

Liu

et al. 2024

Eng. Res. Express

View full text Add to dashboard Cite

A fault distance measurement method that is suitable for long-distance wind power transmission lines is proposed by this paper. By combining the voltage and current at both ends of the line during normal operation, a dynamic line parameter correction method based on improved NSGAⅡ is proposed to ensure measurement errors caused by parameter changes are avoided. Based on the construction of the fault distance measurement function, a global optimization method that is based on adaptive golden section search is proposed, where only the power frequency measurement data at both ends of the bus is required and synchronization is not required for the measurement of fault distance. A large number of experimental results demonstrate that the method this paper proposes is highly adaptable, with a relative distance measurement error of less than 0.2km, and high accuracy is maintained in distance measurement even with certain degrees of noise interference.

show abstract

“…VMD is a technique for signal decomposition and modal analysis that can adapt to the desired signal and perform filtering model correction [ 25 ]. In a study by Sauvik Biswas et al, an intelligent fault detection and classification for UPFC transmission lines and wind farms was developed based on variational modal decomposition–CNN using VMD to decompose and extract optimal signals from locally measured current signals [ 26 ]. The VMD algorithm decomposes the original signal into multiple intrinsic modal functions (IMFs) with different center frequencies.…”

Section: Introductionmentioning

confidence: 99%

“…The VMD algorithm decomposes the original signal into multiple intrinsic modal functions (IMFs) with different center frequencies. This allows for the feature judgment and reconstruction of different frequency components in the signal [ 25 , 26 ].…”

Section: Introductionmentioning

confidence: 99%

Time Delay Study of Ultrasonic Gas Flowmeters Based on VMD–Hilbert Spectrum and Cross-Correlation

Kong,

Zhang,

Guo

et al. 2024

Sensors

View full text Add to dashboard Cite

The accuracy of ultrasonic flowmeter time delay measurement is directly affected by the processing method of the ultrasonic echo signal. This paper proposes a method for estimating the time delay of the ultrasonic gas flowmeter based on the Variational Mode Decomposition (VMD)–Hilbert Spectrum and Cross-Correlation (CC). The method improves the accuracy of the ultrasonic gas flowmeter by enhancing the quality of the echo signal. To denoise forward and reverse ultrasonic echo signals collected at various wind speeds, a Butterworth filter is initially used. The ultrasonic echo signals are then analyzed by Empirical Mode De-composition (EMD) and VMD analysis to obtain the Intrinsic Mode Function (IMF) containing distinct center frequencies, respectively. The Hilbert spectrum time–frequency diagram is used to evaluate the results of the VMD and EMD decompositions. It is found that the IMF decomposed by VMD has a better filtering performance and better anti-interference performance. Therefore, the IMF with a better effect is selected for signal reconstruction. The ultrasonic time delay is then calculated using the Cross-Correlation algorithm. The self-developed ultrasonic gas flowmeter was tested on the experimental platform of the gas flow standard devices using this signal processing method. The results show a maximum indication error of 0.84% within the flow range of 60–606 m3/h, with a repeatability of no more than 0.29%. These results meet the 1-level accuracy requirements as outlined in the national ultrasonic flowmeters calibration regulation JJG1030-2007.

show abstract

An intelligent fault detection and classification technique based on variational mode decomposition-CNN for transmission lines installed with UPFC and wind farm

Cited by 30 publications

References 29 publications

Convolutional Neural Network-Based Pattern Recognition of Partial Discharge in High-Speed Electric-Multiple-Unit Cable Termination

Convolutional Neural Network-Based Pattern Recognition of Partial Discharge in High-Speed Electric-Multiple-Unit Cable Termination

Fault distance measurement method for wind power transmission lines based on improved NSGA II

Time Delay Study of Ultrasonic Gas Flowmeters Based on VMD–Hilbert Spectrum and Cross-Correlation

Contact Info

Product

Resources

About