基于多维关联信息的电压暂降治理需求识别  

Identify the Mitigation Demand Against Voltage Sag Based on Multidimensional Related Information

作　　者：汪颖[1] 王曼 陈韵竹 肖先勇[1] 刘可 车克杉 WANG Ying;WANG Man;CHEN Yunzhu;XIAO Xianyong;LIU Ke;CHE Keshan(College of Electrical Engineering,Sichuan University,Chengdu 610065,Sichuan Province,China;State Grid Qinghai Electric Power Research Institute,Xining 810000,Qinghai Province,China)

机构地区：[1]四川大学电气工程学院,四川省成都市610065 [2]国网青海省电力公司电力科学研究院,青海省西宁市810000

出　　处：《电网技术》2022年第11期4391-4402,共12页Power System Technology

基　　金：国家自然科学基金项目(52077145)。

摘　　要：准确识别用户的电压暂降治理需求,有助于指导用户的电压暂降治理精准投资决策,保证电网公司优质供电服务的顺利推进。基于用户的多维关联信息,提出用户电压暂降治理需求识别模型。首先,构建用户电压暂降治理需求特征库,包括特征指标定义、条件特征和结果特征的计算方法。其次,对连续特征进行离散化处理,筛选出用于需求识别的最优特征,建立需求识别决策表。然后,提出自适应频繁模式增长(frequent-pattern growth,FP-Growth)算法提取关联规则,并提出改进的余弦相似度方法匹配关联规则,识别用户电压暂降治理需求。最后,基于60个中小型工业用户数据,验证了所提模型的有效性与正确性。It is important to identify the mitigation demand against voltage sag(IMD-VS) of the users, guiding the investment decisions of IMD-VS accurately and ensuring the premium power supply services smoothly going well. This paper proposes the IMD-VS model of users based on the users’ multidimensional related information. Firstly, this paper constructs the characteristic database of IMD-VS, including presenting the definitions of characteristics indexes, the calculation methods for the condition characteristics and result characteristics indexes. Secondly, this paper discretizes the continuous characteristics to select the optimal characteristics for IMD-VS, establishing the IMD-VS decision table.Furthermore, this paper proposes an adaptive FP-growth algorithm to extract the association rules, and presents an improved cosine similarity method to match the association rules to identify mitigation demand against voltage sag for the users. Finally, the proposed model is verified by the data from 60 small and medium-sized industrial users, the effectiveness and correctness of the proposed model verified.

关 键 词：电压暂降 需求识别 最优特征 关联规则 自适应FP-Growth算法 

分 类 号：TM721[电气工程—电力系统及自动化]