基于改进层次分析法和数据挖掘的架空输电线路极端灾害风险评估模型

An Extreme Disaster Risk Assessment Model for Overhead Transmission Lines Based on Improved Analytic Hierarchy Process and Data Mining

摘要:

针对现有架空输电线路极端灾害风险评估模型仅计算了风险指标的主观权重值导致评估效果不佳的问题，提出了一种基于改进层次分析法和数据挖掘的评估模型。首先，从输电线路的故障隐患、运行状况和自然条件3个方面出发，建立风险评估体系层次结构；然后，利用层次分析法将相同层次的不同风险指标进行两两比较，得到各风险指标的主观权重值，再利用最小相对熵原理对主观权重值进行改进计算得到风险指标的客观权重值;最后，利用数据挖掘技术从历史极端灾害数据中提取出各风险指标的风险特征及风险等级，生成风险评估模型。通过实验对比测试，所提出的基于改进层次分析法和数据挖掘的架空输电线路极端灾害风险评估模型评估准确率为96.7%，评估效果较好。

Abstract:

Previous models for extreme disaster risk assessment of overhead transmission lines only calculated the subjective weight values of risk indicators, resulting in poor evaluation performance. Therefore, an extreme disaster risk assessment model for overhead transmission lines based on improved analytic hierarchy process (AHP) and data mining is designed. Firstly, based on the actual situation of overhead transmission lines, a hierarchical structure of risk assessment system is established from three aspects: fault hidden dangers, operating conditions and natural conditions of transmission lines. And then, by comparing different risk indicators at the same level in pairs, subjective weight value of risk indicators is obtained. The subjective weight value of risk indicators is calculated and improved with the principle of minimum relative entropy to obtain the objective weight value of risk indicators. Finally, data mining technology is used to extract the corresponding risk features and risk levels form historical extreme disaster data, and the corresponding risk assessment model is generated. Through experimentalcomparison testing, the proposed model based on improved AHP and data mining of extreme disaster risk assessment for overhead transmission lines has an accuracy of 96.7%, and the evaluation effect is good.