Fault arc detection method and detection apparatus

Abstract

The invention discloses a fault arc detection method and detection apparatus, and relates to detection of electrical faults. The method comprises the following steps: acquiring circuit signals; determining whether an acquisition point quantity reaches a preset value of an acquisition point quantity within 0.5 s, if so, performing the next step, and otherwise, returning; performing wavelet threshold noise reduction on the acquired signals by use of DB4 wavelets; performing two-layer wavelet transformation on the signals after the noise reduction; extracting a second-layer high-frequency coefficient; processing the second-layer high-frequency coefficient; counting the quantity of points which are not zero in the processed second-layer high-frequency coefficient; and if a statistical value is greater than a preset threshold, giving an alarm, and otherwise, returning to the first step. A single-chip microcomputer module is connected with a signal conditioning module, a driving module and a power supply module, the signal conditioning module is connected with a sensor module, and the single-chip microcomputer module comprises a signal acquisition module and a calculation module. According to the invention, the purposes of improving fault arc detection precision, reducing the false alarm rate and improving the detection accuracy are realized.

Description

technical field [0001] The invention relates to a detection device for electrical faults, in particular to detection of faults in cables, transmission lines or networks, in particular to a fault arc detection method and detection device. Background technique [0002] Fault arc detection technology is to convert harmful arc pulses (such as arcs caused by wire insulation drop, load short circuit, and poor contact of switch connectors) that are prone to electrical fires from a large number of harmless arc pulses (such as energy-saving , electric hair dryer, electric hand drill, vacuum cleaner, arc welding machine, air compressor, high-power switch action, etc.) and interference pulses are detected, and then the power is cut off without malfunctioning. There are many methods for detecting arc faults in the power grid, and some algorithms require too many thresholds, which are difficult to coordinate; the setting of thresholds depends heavily on experience and data, and is diffic...

AI Technical Summary

Problems solved by technology

There are many methods for detecting arc faults in the power grid, and some algorithms require too many thresholds, which are difficult to coordinate; the setting of thresholds depends heavily on experience and data, and is difficult to promote

In some methods, the initial normal waveform data needs to be input in advance. If the load is different, it may be falsely reported; the normal waveform data is updated dynamically. If the waveform does not exceed the threshold every time it is updated, it will be accumulated multiple times. Under this condition, the data of the normal waveform will no longer be normal, making it impossible to detect

It also requires multiple thresholds, depends on data and experience, and is difficult to promote

In some methods, the typical value of the high-order harmonic content rate will be updated dynamically. If there is a change that does not exceed the threshold value in each update, the typical value will no longer be normal after multiple accumulations, so it cannot be detected.

The criterion is too single, and it is easy to misdetect only relying on the detection of high-order harmonics

For some methods, when the load power is small, it cannot be detected

Images