A method, device and system for arc fault detection based on probabilistic neural network

Abstract

The invention discloses a fault arc detection method, device and system based on a probabilistic neural network. The method includes: training set preprocessing: collecting arc waveform continuous signals of fault arcs and normal arcs under different loads as a continuous signal sample set; Carry out ADC sampling on the continuous signal sample set to obtain the discrete signal sample set; perform PCA dimension reduction processing on the discrete signal sample set to obtain the low-dimensional discrete signal sample set; train the probabilistic neural network: use the low-dimensional discrete signal sample set as the probability neural network inputting, constructing a probabilistic neural network framework, and obtaining a training probabilistic neural network model; the training probabilistic neural network model can be used for fault arc detection. The invention can quickly establish a training model, accurately analyze and judge fault arcs in real time, has good anti-interference and noise capabilities, and greatly reduces the occurrence of false alarms.

Description

technical field [0001] The invention relates to the field of arc fault detection, in particular to a method and device for arc fault detection based on a probabilistic neural network. Background technique [0002] At present, the detection algorithm of arc fault in China is still widely in the research stage. The common identification methods are mostly based on the electrical characteristics of arc, including light, heat, current change rate, wavelet frequency band, etc. However, when there are many types of loads, the identification becomes more difficult. Due to the limit of the line load, the fault current is so small that the existing system cannot realize the protection of the series arc fault, and there are potential electrical safety hazards. When the arc occurs, it will generate a lot of heat, which can easily ignite the surrounding combustibles, thus causing a fire. Compared with traditional electrical faults, the current amplitude of fault arc changes less, and t...

AI Technical Summary

Problems solved by technology

[0004] The technical problem to be solved by the present invention is to provide a fault arc detection method with high discrimination accuracy and low false alarm rate, so as to solve the problem that it is difficult to accurately detect series arc faults due to the limitation of line load in the existing fault arc detection

Images