Circuit breaker fault arc detection method based on VMD parameter optimization and sample entropy

Abstract

The invention discloses a circuit breaker fault arc detection method based on VMD parameter optimization and sample entropy, and the method comprises the following steps: S1, collecting an arc current signal of a fault line, determining a wavelet basis function and a corresponding decomposition level N through the actual electrical characteristics of the signal, and calculating an optimal wavelet packet decomposition tree of the signal according to the existing standard entropy; s2, decomposing the fault arc current of the circuit breaker by using a VMD method, and performing parameter optimization on the decomposed arc current parameter; and S3, calculating the sample entropy SE of each modal component obtained through decomposition in the step S2, and inputting the obtained fault arc signal feature vector into the SVM for training and testing, thereby achieving the purpose of fault arc identification of the intelligent low-voltage circuit breaker. According to the method, the signal features are extracted through the parameter-optimized VMD algorithm, the effect is better, and the accuracy is higher by inputting the feature vectors obtained by calculating the sample entropy SE into the SVM for fault arc recognition.

Description

technical field [0001] The invention relates to the technical field of electrical engineering, in particular to a circuit breaker fault arc detection method based on VMD parameter optimization and sample entropy. Background technique [0002] In the low-voltage distribution network, in order to reduce the harm to residents and industrial users, the development and application of arc fault detection circuit breakers are very important. When a fault arc occurs, a lot of information can be analyzed by detecting the current signal in the power supply network, which has a great effect on fault discrimination. [0003] The existing algorithms for detecting arc faults in low-voltage circuit breakers also extract and analyze the fault current, and it is crucial to choose which signal processing method to analyze the arc fault. The traditional methods used in the analysis and processing of arc fault signals include Fourier transform, wavelet transform and other time-domain analysis ...

AI Technical Summary

Problems solved by technology

Fourier transform is a global analysis, which has limitations in describing the time-frequency local characteristics of the signal; wavelet transform will be affected by the harmonics in the vicinity, and the signal analysis will be distorted

At the end of the 20th century, the Empirical Mode Decomposition (EMD) method was proposed to analyze nonlinear and adaptive signals. It decomposed the fault arc signal to obtain multiple IMF components, but its defect was that the decomposition The effect is not good, it will produce modal aliasing, endpoint effect, and redundant components

[0004] At present, this research method has been applied to the fault diagnosis of the low-voltage circuit breaker itself, but it is limited to the decomposition of the vibration signal when the mechanical failure of the circuit breaker occurs.

Images