Fault arc multi-domain identification method based on KPCA-MIV-LSTM

Abstract

The invention relates to a fault arc multi-domain identification method based on KPCA-MIV-LSTM. The method comprises the following steps: acquiring single-phase current signals under different load conditions; non-correlation components in the measurement signals are separated, and the obtained single-phase current is preprocessed; selecting a first principal component of which the contribution degree exceeds 85% in the signal preprocessed by the KPCA algorithm as a subsequent processing target to carry out time domain, frequency domain and energy domain multi-domain feature analysis; evaluating and selecting high-correlation characteristics under the corresponding load condition; and taking the screened high-correlation features as a training set and a test set of the LSTM long short-term memory network, and carrying out detection and identification on series arc faults under different load conditions. The method comprises the following steps of: performing KPCA (Kernel Principal Component Analysis) on a fault series arc current signal, separating a non-correlation component in a measurement signal, and selecting a first principal component of a solution with an optimal contribution degree; and analyzing and processing the signals after KPCA preprocessing from various indexes in a time domain, a frequency domain and an energy domain.

Description

technical field [0001] The invention relates to the technical field of identification of series fault arcs under low-voltage levels, in particular to a multi-domain identification method of fault arcs based on KPCA-MIV-LSTM. Background technique [0002] The current arc fault detection technology has the following four categories, and their respective defects are as follows: [0003] (1) Arc fault detection technology based on simulation model [0004] Previously, a large number of scholars have studied the simulation of arc models, and derived various arc models based on the law of energy conservation and arc column plasma characteristics, such as the Cassie model for simulating high-resistance arcs and the Mayr model for simulating low-resistance arcs. And based on Mayr's improved Shavemaker model, etc. Although the establishment of the arc model can simulate the arc waveform to obtain the arc characteristics to a certain extent, and the cost is low, there are some assum...

AI Technical Summary

Problems solved by technology

[0011] The purpose of the present invention is to provide a KPCA-MIV-LSTM-based fault arc multi-domain identification that can realize effective identification of series arc faults at low voltage levels, reduce the amount of calculation and improve detection speed and accuracy, and solve the difficult problem of identifying series arc faults method

Images