基于系统动态机理差异度的广域电力系统负荷模型自适应优化研究

项目名称： 基于系统动态机理差异度的广域电力系统负荷模型自适应优化研究

项目编号： No.51307078

项目类型： 青年科学基金项目

立项/批准年度： 2014

项目学科： 电工技术

项目作者： 郝丽丽

作者单位： 南京工业大学

项目金额： 25万元

中文摘要： 广域电力系统负荷面广量大，其准确建模对电力系统仿真的有效性至关重要，通常根据负荷特征分类识别，但却较少考虑系统运行状态变化及负荷自身时变性的影响。项目选择广域电力系统负荷为研究对象，以提高系统整体动态仿真有效性和负荷模型适用性为目标，重点研究能够跟踪系统实际工况和负荷变化的广域系统负荷模型参数自适应优化方法。基于系统功角、电压稳定特征建立电力系统受扰轨迹的机理差异度，用其作为广域系统负荷模型识别的目标函数。根据负荷对失稳模式的影响差异，建立负荷的稳定群属特征，并协同负荷构成比例特征研究广域系统负荷节点分类策略。运用统计和聚类技术建立各节点的负荷特征索引表，根据实际运行环境实时匹配得到各节点的负荷构成比例特征，根据系统实际运行状态判断各节点的稳定群属特征，据此对负荷分类并动态优化各类参数。项目为广域电力系统负荷可靠、实用模型的建立提供了有力的分析方法，为电力系统安全、经济运行提供了保障。

中文关键词： 系统动态的机理差异度；广域系统负荷识别；分类策略；时变性；自适应优化

英文摘要： For multitude and broad distribution of loads, the accurate electric load modeling of wide-area power system is essential to validity of power system simulation. Usually the models are identified after classification according to the load characteristics, but the influence of system working state changes and load time variation is rarely considered. The loads of wide-area power system are selected as the research object, this item aims at the improving of validity for overall system dynamic simulation and applicability of load models. The research focus on the adaptive optimization method for wide-area sytem load parameters to track the actual system conditions and load changes. The mechanism difference degree of power system disturbed trajectories, which is established based on power angle and voltage stability characteristic of power system, can be used as the objective function to identify wide-area system load models. According to the different influence of loads on system unstable mode, the stability clustering characteristic of load is established, based on which the load classification strategy for wide-area system is studied together with load composition proportion characteristic. The load characteristic index table is established using statistics and clustering technology, which is used to match with t

英文关键词： mechanism difference degree of system dynamic；wide-area system load identification；classification strategy；time variation；adaptive optimization