Streaming data fast distribution method and system for online ultra-real-time simulation of electric power system

Abstract

The present invention proposes a streaming data fast distribution method and system for online ultra-real-time simulation of a power system, and uses bidirectional bandwidth, thread-level parallelism, and pipeline technology to realize mutual parallel transmission of data streams in computing nodes to improve resource utilization. The goal of the present invention is to realize the rapid distribution of streaming data for online ultra-real-time simulation of power systems, and to efficiently distribute 100-megabyte simulation data files to hundreds of nodes in a computing cluster in seconds in a 10-gigabit network environment, thereby achieving The technical effect of online ultra-real-time simulation.

Description

technical field [0001] The present invention relates to the field of streaming data distribution, and more specifically, to a system and method for fast streaming data distribution of online ultra-real-time distributed simulation calculation of electric power system. Background technique [0002] In recent years, the high-performance computing cluster management technology has developed rapidly, and the use of high-performance computing clusters for power system parallel simulation calculations has become a reality, playing a key supporting role in the field of power system offline simulation and online analysis. However, as the interconnection scale of my country's power system continues to expand, in order to adapt to the wide application of new devices and technologies and the changing characteristics of the operation mode after the power marketization, and to further improve the safe and stable operation level of the power system, it is necessary to realize the power syst...

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to solve the problems of low efficiency, unreliability and poor scalability in the distribution of large-scale data in the prior art, and proposes a streaming data distribution system for online super-real-time simulation of power systems. This platform adopts a The hierarchical streaming data distribution method based on the reliable transmission protocol, using thread-level parallel technology and pipeline technology, realizes the one-to-many distribution of real-time data of large power grid simulation calculations to all available nodes of the cluster, improving reliability while ensuring reliability. distribution efficiency

Images