基于图划分的大电网拓扑分析

为支撑大电网在线实时监控和分析，提出了一种基于图划分的大电网拓扑分析方法。该方法首先根据智能电网调度控制系统电网建模特点，给出了基于图论的电网层次结构、拓扑结构、厂站母线分析和系统网络分析模型。然后基于图划分建立了并行网络拓扑和局部拓扑修正的数学模型，并根据该模型设计了大电网并行网络拓扑的实现方法。最后基于共享内存编程模型在智能电网调度控制系统中研发了快速网络拓扑分析服务功能。所提出的模型和方法实现了大电网并行网络拓扑分析的无锁计算，避免了多线程数据竞争导致的阻塞耗时问题。对实际系统进行仿真测试，结果表明了该方法的准确性、实时性和有效性。

Large power grid topology analysis based on graph partitioning

In order to support the online real-time monitoring, control and analysis of large power grid, this paper proposes a new topology analysis method for large power grid based on graph partitioning. According to the model characteristics of the smart grid dispatching and control systems, this method first provides the power grid model of hierarchical structure, topology structure, the bus of substation analysis and system network topology. Then it establishes mathematical model for parallel network topology and local topology modification based on graph partitioning. According to this model, the realization method of the large power parallel topology analysis is designed. Finally using shared memory programming model, the function of the large power fast topology analysis services is developed in the smart grid dispatching and control systems. The models and method proposed realizes the no lock computing of large power grid parallel topology analysis, and avoids the blocking time-consuming problem caused by multithreaded data competition. Simulation is carried out by taking the actual power grid as an example, the veracity, feasibility and practicality of the proposed method are verified. This work is supported by Science and Technology Project of State Grid Corporation of China (No. KY-SG-2016- 031-JLDKY).