Abstract
To further improve the detection ability of residual current in low-voltage distribution networks, an adaptive residual current detection method based on variational mode decomposition (VMD) and dynamic fuzzy neural network (DFNN) is proposed. First, using the general <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$K$ </tex-math></inline-formula> -value selection method of VMD proposed in this study, the residual current signal is decomposed into <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$K$ </tex-math></inline-formula> intrinsic mode functions (IMFs). By introducing the cross-correlation coefficient <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$R$ </tex-math></inline-formula> and the time-domain energy entropy ratio <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$E$ </tex-math></inline-formula> as two classification indexes, IMFs are divided into three categories: effective IMFs, noise IMFs and aliasing IMFs. Then, the aliasing IMFs are denoised by recursive least squares (RLS), and the denoised IMFs are superimposed with the effective IMFs to obtain the reconstructed signal. Finally, the dynamic fuzzy neural network (DFNN) is adjusted by the minimum output method to achieve the detection of the reconstructed residual current signal, and the network is used to predict the residual current according to the detection results. The detection results of the simulation and measured data show that the proposed algorithm has high detection accuracy and is superior to the wavelet neural network, empirical mode decomposition-thresholding, and wavelet entropy-auto encoder-back propagation neural network methods in terms of mean square error, goodness of fit and running time. This method provides a reference for further research on new adaptive residual current protection devices.
Highlights
With the widespread application of electrical equipment, it has been difficult for traditional residual current devices (RCD) has been difficult to meet the complex electrical environment and the requirements of the public for daily electrical safety
To solve the problem of residual current detection in lowvoltage distribution networks, we proposed a residual current detection method based on RE-recursive least squares (RLS)-variational mode decomposition (VMD) and dynamic fuzzy neural network (DFNN)
In order to solve the problem that the K -value of VMD is difficult to determine when processing the residual current signal, we proposed the concept and determination method of general K -value, which reduces the operation time and operation volume compared with the method of seeking the optimal K -value
Summary
With the widespread application of electrical equipment, it has been difficult for traditional residual current devices (RCD) has been difficult to meet the complex electrical environment and the requirements of the public for daily electrical safety. Wang proposed a matrix transformation method to deal with the double summation inequality with fuzzy weight functions, which has good stability [15] These methods have optimized the structure and parameters of neural networks, and have achieved good results in different applications. The number of VMD decomposition levels K needs to be determined in advance, so we proposed the concept and selection method of the general K -value This method can preliminarily determine the K -value of the residual current for a certain fault type, which avoids a large number of calculations caused by seeking the optimal K -value and improving the adaptive ability of VMD. To solve the problem of inaccurate information extraction caused by the aliasing effect that may exist in VMD, we proposed a classification method of IMFs with cross-correlation coefficient R and time domain energy entropy ratio E as the classification indexes [17]–[19].
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
