基于多Agent系统的大型电网电压暂降评估

面向实际大型电网的暂降评估需要考虑覆盖全网特征的线路故障仿真，其仿真计算速度和稳定性是制约其实用化的主要因素。为此提出了一种基于多Agent系统的大型电网电压暂降评估方法。该方法根据仿真需求构造多Agent系统并制定协同交互规则。首先，由网架解析和故障设置Agent自主生成仿真任务。其次，由仿真管理和计算Agent动态执行海量仿真。最后，由数据处理和分析Agent异步存储仿真数据并生成评估结果。目前该方法已应用在某省级电网电压暂降评估工作中，取得了较好效果。实例证明所提方法加速比高，运行稳定，可为大型电网电压暂降评估实用化提供技术支撑。

Voltage sag assessment of a large power grid based on a multi-Agent system

Line fault simulation which covers the whole network should be considered in actual large-scale power grid sag assessment. The calculation speed and stability of the simulation are the main factors that restrict its practicability. Therefore, this paper proposes a voltage sag assessment method for a large-scale power grid based on a multi-Agent system. This method constructs the multi-Agent system and formulates collaborative interaction rules according to simulation requirements. First, the simulation task is generated independently by the grid analysis and a fault setting Agent. Second, massive simulation is dynamically executed by simulation management and a computational Agent. Finally, the simulation data are stored asynchronously by the Agent and the evaluation results are generated. At present, this method has been applied to the voltage sag assessment of a provincial power grid, and good results have been achieved. Examples show that this method has high acceleration ratio and stable operation, and can provide technical support for the practical application of voltage sag assessment of a large power grid.