基于自学习权重-云模型的工业用户外界风险评估方法  

External Risk Assessment Method for Industrial Users Based on Self-learning Weight-cloud Model

作　　者：陈宗汉 李明 徐永海[1] 何胜 袁金斗 杨恒 潘明明 CHEN Zonghan;LI Ming;XU Yonghai;HE Sheng;YUAN Jindou;YANG Heng;PAN Mingming(School of Electrical and Electronic Engineering,North China Electric Power University,Beijing 102206,China;State Grid Corporation of China,Beijing 100031,China;Beijing Key Laboratory of Demand Side Multi-Energy Carriers Optimization and Interaction Technique,China Electric Power Research Institute Co.,Ltd.,Beijing 100192,China)

机构地区：[1]华北电力大学电气与电子工程学院,北京102206 [2]国家电网有限公司,北京100031 [3]需求侧多能互补优化与供需互动技术北京市重点实验室(中国电力科学研究院有限公司),北京100192

出　　处：《华北电力大学学报(自然科学版)》2025年第2期80-90,共11页Journal of North China Electric Power University:Natural Science Edition

基　　金：国家电网有限公司总部管理科技研究项目(5108-202218280A-2-0-XG).

摘　　要：针对传统权重评估方法存在的缺陷,以及面对复杂多变电网环境时工业用户用电安全潜存风险的问题,提出一种基于自学习权重-云模型的工业用户外界风险定级评估方法。首先从节点电压变动、削减量与可调控量占比、线路过载、孤岛问题和节点电压谐波5个角度构建了工业用户外界风险评估指标体系;其次利用基于区间估计的AHP法(Analytic Hierarchy Process,AHP)、熵权法和独立性权法分别计算评估指标的主观、客观和独立性权重,并提出基于权重相对变化率准则的自学习权重模型得到组合权重,增强评估方法动态适应性。基于风险指标后果的严重程度改进云模型风险评估矩阵构建工业用户外界风险定级评估方法;最后,通过基于工业用户外界风险评估指标体系,以IEEE-RTS-24系统为实际场景验证了该风险定级评估方法的合理性。Given the defects of the traditional weight assessment method and the potential risks to industrial users’electricity safety in the face of a complex and changing grid environment,we propose a risk rating assessment method for industrial users’external risks based on self-learning weight-cloud model.Firstly,we construct an industrial user external risk assessment index system from five perspectives:node voltage variation,reduction and adjustable volume ratio,line overload,islanding problem and node voltage harmonics.Secondly,we calculate the subjective,objective and independent weights of the assessment indexes using the Analytic Hierarchy Process(AHP)method based on interval estimation,the entropy weight method and the independence weight method,respectively.Then,we propose a self-learning weight model based on the criterion of the relative rate of change of weights to obtain the combined weights and enhance the dynamic adaptability of the assessment method.Furthermore,based on the severity of the consequences of the risk indicators,we improve the cloud model risk assessment matrix to establish a risk rating assessment method for industrial users.Finally,the rationality of the proposed risk rating assessment method is verified by using the IEEE-RTS-24 system as a practical scenario,based on the external risk assessment indicator system for industrial users.

关 键 词：自学习综合权重 云模型 工业用户 风险评估 

分 类 号：TM76[电气工程—电力系统及自动化]