Self-learning multifunctional fault arc detector

Abstract

The invention discloses a self-learning multifunctional fault arc detector. The device comprises a core device and a switching power supply, the core device is connected with the switching power supply; the core device comprises a core processor, a data storage unit, an RS485 communication unit, a wireless communication unit, a relay output unit, an HMI human-computer interface, a case input unit,a case signal conditioning unit, a voltage and current signal sampling unit, an electric leakage signal sampling unit and a temperature signal sampling unit, wherein the core processor is connected with the data storage device, the RS485 communication device, the wireless communication device, the relay output device, the HMI human-computer interface, the case input device and the case signal conditioning device, and the case signal conditioning device is connected with the voltage and current signal sampling device, the electric leakage signal sampling device and the temperature signal sampling device. The method has the beneficial effects that by collecting the waveform slope and the high-frequency characteristic of the current and calculating whether the fault arc exists or not throughthe BP neural network, a user is helped to more intuitively know the power consumption condition of the power line, and the cause of the fault arc is more accurately positioned.

Description

technical field [0001] The present invention relates to the technical field of arc field, in particular to a self-learning multifunctional arc fault detector. Background technique [0002] Traditional arc fault detectors rely on arc light, arc sound and temperature to judge whether there is an arc. Due to the uncertainty of the location of the fault arc in the power line, the above detection methods and devices based on the physical characteristics of the arc are not suitable for faults in the power line. Arc detection. [0003] In order to improve the accuracy of arc fault detection, a multi-feature fusion arc fault algorithm is needed. According to the true effective value, flat shoulder characteristics, high frequency characteristics and waveform slope characteristics of the current waveform, the BP neural network algorithm is used to judge the arc fault. . At the same time, in order to make up for the unreliability of on-site arc fault judgment due to the single sample...

AI Technical Summary

Problems solved by technology

[0002] Traditional arc fault detectors rely on arc light, arc sound and temperature to judge whether there is an arc. Due to the uncertainty of the location of the fault arc in the power line, the above detection methods and devices based on the physical characteristics of the arc are not suitable for faults in the power line. Arc Detection

Images