Node transfer network partitioning optimization method based on rapid splitting algorithm

Abstract

The invention discloses a node transfer network partitioning optimization method based on a rapid splitting algorithm. The method comprises steps of selecting a weak line based on the rapid splitting algorithm; transferring nodes connected to a current cutting branch from the interior of a corresponding subnetwork to the interior of a coordination network according to a certain principle, namely, a tearing node set; and defining a cutting object conditional function K, representing the cutting rationality with a formula described in the specification, wherein the symbol "|" represents a precondition, the right of the symbol is the prediction, the left of the symbol is the final object conditional function value, Vconet is the number of the tearing nodes included in the coordination network, and Vnet_i and Vnet_j are the total numbers of nodes included in the ith and jth subnetworks after splitting. When being applied to parallel calculation of the electric power system trend, the method is more comprehensive and systematic, and provides an objective basis for determination of splitting node members of the node splitting algorithm.

Description

technical field [0001] The invention relates to the block level of a power network, in particular to a block optimization method for a node migration network based on a fast splitting algorithm. Background technique [0002] The basic idea of ​​power system parallel computing is to divide the power network into several sub-networks and a coordination network according to the geographical or network topology connection structure, and the sub-networks are connected with each other through a set of boundary quantities corresponding to the coordination network. At this time, parallel calculations can be implemented for the internal correlation operations of the sub-networks. Among them, the node splitting method, as one of the splitting methods, has the advantage of dividing the power system network to achieve parallel computing, but at the same time, it cannot objectively and reasonably determine the splitting node set, resulting in a bottleneck that cannot be extended. Conte...

AI Technical Summary

Problems solved by technology

Among them, as one of the segmentation methods, the node splitting method has the advantage of dividing the power system network to achieve parallel computing, but at the same time, it will not be able to objectively and reasonably determine the split node set, resulting in a bottleneck that cannot be expanded.

Images