Novel power distribution network fault detection method based on quick switch and fault self-healing system

Abstract

The embodiment of the invention provides a novel power distribution network fault detection method based on a quick switch. The method comprises the steps of fault current signal acquisition, data extraction, neural network training, fault classification and quick switch action control. A multi-layer sensor is adopted to acquire current information, and a particle swarm optimization algorithm is adopted to diagnose hierarchical information. A line E-type traveling wave distance measurement method is adopted to estimate the position of the fault, and is combined with an Attention mechanism to judge the position of the fault. An eddy current repulsion is adopted as a quick switch of a driving mechanism, and breaking can be achieved within 3 ms after a fault action signal is received. The method improves the fault detection level, improves the utilization rate of power supply equipment, reduces the electric energy loss, and improves the overall operation level of a power distribution network. The fault self-healing capability of the power distribution network is improved, and the power supply reliability is effectively improved. The embodiment of the invention further provides a novelpower distribution network fault self-healing system based on the method.

Description

technical field [0001] The invention belongs to the technical field of distribution network fault detection and isolation, and in particular relates to a novel distribution network fault detection method based on a fast switch. The invention also provides a fault self-healing system applying the distribution network fault detection method. Background technique [0002] With the development of modern intelligent power distribution technology, the scale of distribution network construction is getting larger and larger. It has evolved from traditional passive distribution network to active distribution network including distributed power. There are many points, long lines, wide areas, and harsh operating conditions, so single-phase ground faults and short-circuit faults are prone to occur. The penetration of distributed power sources leads to two-way flow of power flow in the distribution network, and creates new problems for the distribution network, such as fault misjudgment...

AI Technical Summary

Problems solved by technology

[0008] Aiming at the problems of the existing distribution network fault detection technology, such as long detection time, unclear fault type, high misjudgment rate, and poor fault tolerance, the present invention provides a distribution network fault detection method and fault self-healing system based on neural network and fast switch , improve the fault detection level and fault self-healing ability of the distribution network, and improve the overall operation level and power supply reliability of the distribution network

Images