基于改进K-means与组合赋权的智能变电站二次系统信息质量评价  

Information Quality Evaluation of Smart Substation Secondary System Based on Improved K-means and Combined Weighting

作　　者：孟令雯 高吉普 张庆伟 辛明勇 席禹 MENG Lingwen;GAO Jipu;ZHANG Qingwei;XIN Mingyong;XI Yu(Electric Power Research Institute,Guizhou Power Grid Co.,Ltd.,Guiyang 550000,Guizhou Province,China;Guiyang Power Supply Bureau,Guizhou Power Grid Co.,Ltd.,Guiyang 550000,Guizhou Province,China;China Southern Power Grid Digital Grid Research Co.,Ltd.,Guangzhou 510000,Guangdong Province,China)

机构地区：[1]贵州电网有限责任公司电力科学研究院,贵州省贵阳市550000 [2]贵州电网有限责任公司贵阳供电局,贵州省贵阳市550000 [3]南方电网数字电网研究有限公司,广东省广州市510000

出　　处：《电力信息与通信技术》2023年第5期53-60,共8页Electric Power Information and Communication Technology

基　　金：南方电网公司重点科技项目“智能变电站二次系统信息治理关键技术研究与应用”(GZKJXM20191312)。

摘　　要：随着变电站的智能化和信息化水平突飞猛进,变电站产生和存储的数据越来越多,数据异常和数据缺失等数据质量问题逐渐突出。作为提高数据质量的基础和前提,智能变电站二次系统数据质量评价尤为重要。文章提出一种基于改进K-means和组合赋权法的智能变电站二次系统信息质量评价方法,该方法首先通过正确性、完善性等6个指标对智能变电站二次系统的信息多层次、全方位分析,解决传统智能变电站二次系统信息质量评价指标单一、评价体系简单的问题;其次,针对智能变电站二次系统数据处理的实时性以及准确性的问题,采用改进K-means算法对数据进行预处理;再次,采用变异系数法与序关系法对评价指标进行组合赋权,使权重既带有客观因素,也包含专家的主观判断,从而使评价结果真实可靠;最后,以仿真数据为例说明该方法的有效性。With the rapid advancement of substation intelligence and informatization,more and more data are generated and stored in substations,and data quality problems such as data abnormalities and data missing are gradually becoming prominent.As the basis and premise of improving data quality,the data quality evaluation of smart substation secondary system is very important.This paper proposes an information quality evaluation method for the secondary system of smart substations based on improved K-means and combined weighting.First,this method analyzes the information of the secondary system in a multi-level and omni-directional manner through six indicators such as correctness and completeness,and solves the problems of single information quality evaluation index and simple evaluation system of the traditional smart substation secondary system.Secondly,for the real-time and accuracy of the data processing of the smart substation secondary system,the improved K-means algorithm is used to process the data.Thirdly,this paper use the coefficient of variation method and the order relationship method to obtain combine weights for the evaluation indicators,so that the weights contain both objective factors and subjective judgments of experts and the evaluation results are true and reliable.Finally,the simulated data is taken as an example to illustrate the method.

关 键 词：智能变电站二次系统 信息质量评价 改进K-means聚类 组合赋权法 

分 类 号：TM63[电气工程—电力系统及自动化]