ANN-based overhead flexible straight power grid fault detection method

Abstract

The invention discloses an artificial neural network (ANN)-based overhead flexible straight power grid fault detection method. The method comprises the following steps of: first, extracting a high frequency component by using wavelet transform, and processing a relation of positive and negative voltage change amounts by using phase mode transformation, so as to acquire an input characteristic quantity of an artificial neural network; further, designing quantities of nerve cells of a hidden layer and an output layer, and training and testing the artificial neural network through massive samples; and finally, realizing fault detection and fault selection functions according to an online detection result of the artificial neural network. The method overcomes defects of setting difficulty andcalculation complexity of a traditional protection algorithm, makes the function of the fault detection algorithm more perfect, has both a fault recognition function and a fault selection function, improves protection reliability, and is not affected by AC fault, running mode and transition resistance.

Description

technical field [0001] The invention belongs to the field of flexible DC power grids, and more specifically relates to an ANN-based fault detection method for overhead flexible DC power grids. Background technique [0002] With the gradual depletion of fossil energy and the increasing pressure on the environment, China and other countries in the world have begun to adjust their energy structure and develop large-scale clean energy. However, the intermittent and random nature of clean energy such as wind energy and solar energy, coupled with the limited ability of the local AC system to absorb new energy, makes wind and solar abandonment often occur. At the same time, my country's clean energy base and load demand are in reverse distribution, and the long-distance grid connection of large-scale new energy and the optimal allocation of resources have become urgent problems to be solved. Using DC transmission technology based on voltage source converters is an effective soluti...

AI Technical Summary

Problems solved by technology

[0007] Aiming at the above defects or improvement needs of the prior art, the present invention provides an ANN-based fault detection method for overhead flexible direct power grids, which solves the problems of difficult selection of thresholds and complex calculations existing in existing protection algorithms, further improves protection performance, and takes into account Fault identification and fault pole selection function, and improve the reliability of protection, not affected by AC fault, operation mode and transition resistance, provide technical support for safe and reliable operation of DC power grid

Images