Series arc fault intelligent circuit breaker

Abstract

The invention discloses a series arc fault intelligent circuit breaker scheme design based on a machine learning algorithm. The circuit breaker is used for single-phase low-voltage power supply linesfor residents, industry and commerce. A current signal of a circuit is acquired by a sensor. Then zero-zone proportion and peak value coefficient characteristics of the signal are extracted from a time domain, inter-harmonic content characteristics of the signal are extracted from a frequency domain, and power spectrum distribution and harmonic power proportion characteristics of the signal are extracted from an energy domain; and then waveforms are classified according to a kurtosis coefficient and a specific inter-harmonic content, the extracted characteristics of the classified waveforms are calculated by adopting a corresponding intelligent algorithm, whether a series arc fault occurs in the circuit or not is determined, and a fault branch is located. The intelligent algorithm and parameters are obtained by training a machine learning algorithm with a sample and then extracting a rule between the parameters and an algorithm result, and are solidified in a microprocessor. Accordingto the scheme, the judgment accuracy of the arc fault circuit breaker is improved, fault branch positioning can be achieved, an electrical fire problem is solved, and the practical value is achieved.

Description

technical field [0001] The invention relates to a civil low-voltage switching device (hereinafter referred to as a switching device), in particular to a series arc fault circuit breaker. Background technique [0002] Arc is a gas discharge phenomenon accompanied by strong light and high heat, which can easily cause fire or even explosion under certain circumstances, seriously threatening personal safety and electrical equipment safety. When the electrical contact pair in the line connection, connector or switchgear is in poor contact, there is a high probability of inducing a discharge arcing phenomenon between the two electrodes of the contact pair, which is called a series arc fault. Low-voltage switching devices such as miniature air switches (hereinafter referred to as air switches) generally adopt the principle of inverse time current protection, and only trip when the line current meets the overload condition. However, the fault current generated by series arc faults ...

AI Technical Summary

Problems solved by technology

However, the current existing series arc fault detection schemes based on machine learning algorithms all need to use professional graphics workstations to implement machine learning algorithms, which makes such algorithms unable to be applied to actual products

Images