基于停电事件分析的区域性话务峰涌预测  

Prediction of Regional Traffic Rush Based on Blackout Event Analysis

作　　者：李玮 李树国 喻玮 LI Wei;LI Shu-guo;YU Wei(State Grid Co.Ltd.Customer Service Center,Tianjin 300000 China)

机构地区：[1]国家电网有限公司客户服务中心,天津300000

出　　处：《自动化技术与应用》2024年第4期9-13,共5页Techniques of Automation and Applications

摘　　要：研究基于停电事件分析的区域性话务峰涌预测方法,避免停电事件造成的区域性话务峰涌,影响电力用户的体验感。选取Hurst指数作为分析电网停电事件时间序列相关性的分析指标,确定与停电事件高度相关的因素。将停电事件相关因素作为XG boost算法的输入,利用XG boost算法输出停电区域预测结果。采集所预测停电区域的用户数量、话务接通率等数据,作为最小二乘支持向量机的输入样本,利用最小二乘支持向量机输出停电区域话务量预测结果,设置停电区域话务峰涌期阈值,话务量预测结果高于该阈值时,预测该阶段为区域性话务峰涌期。实验结果表明,该方法可以依据停电事件分析结果,精准预测停电区域的话务峰涌期,为电力企业话务调度与话务决策提供依据。A prediction method of regional traffic rush based on power outage event analysis is studied to avoid regional traffic rush caused by power outage event and affect the experience of power users.Hurst index is selected as the index to analyze the correlation of power grid blackout event time series,and the factors that are highly correlated with blackout events are determined.The power outage event related factors are taken as the input of XG boost algorithm,and the prediction results of power outage area are output by XG boost algorithm.Collection predicted blackout area the number of users,the telephone connection,such as data,as the least squares support vector machine(SVM)of input sample,using the least squares support vector machine(SVM)output power failure prediction of regional traffic,it sets up a power outage a swarm of regional traffic period threshold,the traffic forecasting result is higher than the threshold,predicts the stage for a swarm of regional traffic.The experimental results show that the proposed method can accurately predict the traffic rush period in the power outage area according to the analysis results of power outage events,and provide a basis for the traffic scheduling and decision-making of power enterprises.

关 键 词：停电事件分析 区域性 话务峰涌预测 HURST指数 XG boost算法 支持向量机 

分 类 号：TP274[自动化与计算机技术—检测技术与自动化装置]