基于大数据分析的10 kV配网停电作业时长预测优化研究  

Research on the prediction optimization of power outage duration of 10 kV distribution network based on big data analysis

作　　者：唐瑞伟 栗薇[2] 张震[3] 刘津 高原 TANG Ruiwei;LI Wei;ZHANG Zhen;LIU Jin;Gao Yuan(State Grid Tianjin Electric Power Company Ninghe Power Supply Branch,Tianjin 301500,China;State Grid Tianjin Electric Power Research Institute,Tianjin 300000,China;State Grid TianJin Electric Power Company,Tianjin 300010,China;Baodi Power Supply Branch of State Grid Tianjin Electric Power Company,Tianjin 301800,China)

机构地区：[1]国网天津市电力公司宁河供电分公司,天津301500 [2]国网天津市电力公司电力科学研究院,天津300000 [3]国网天津市电力公司,天津300010 [4]国网天津市电力公司宝坻供电分公司,天津301800

出　　处：《电力大数据》2019年第1期27-34,共8页Power Systems and Big Data

摘　　要：传统的10 kV配网作业停电时长预测主要依赖于前期现场勘查和工作负责人经验估算。随着配网作业手段的不断进步及用户供电可靠性、优质服务要求的不断提高,对10 kV配网预安排停电预测的精细、准确程度也提出了较高的要求。针对目前停电时长预测与实际完成偏差仍然较大情况,本文提出了一种基于大数据分析的10 kV配网停电作业时长预测方法。通过开展近三年配网停电事件分析,归纳整理影响预测准确度的不同因素,从标准化作业时长典型基准库、标准化作业时长优化调整机制、标准化作业时长后评估体系以及综合停电时长预测模型四个方面展开深入研究,优化提升综合停电作业时长准确预测能力,为停电作业预测偏差管控提供一个可参考、易操作的高效辅助工具。生产作业实际应用表明,这种基于大数据分析的10 kV配网停电作业计划时长预测方法,可以较好的提升停电作业时长准确预测能力,降低可靠性预测偏差,发挥可靠性目标管控作用,助力停电信息发布等优质服务工作精准开展。The traditional prediction of power outage of operation time in 10 kV distribution network mainly depends on the early site investigations and experience estimation of the person in charge of the work. With the continuous progress of distribution network operation means and the continuous improvement of power supply reliability and quality service requirements of users,higher requirements are put forward for the fine and accurate prediction of 10 kV distribution network outage. In view of the large deviation between the current outage time prediction and the actual completion,this paper presents a method on the prediction of power outage duration of 10 kV distribution network based on big data analysis. Through the analysis of power outages in the past three years,the different factors affecting the accuracy of prediction are summarized,from the typical benchmark library of standardized operation time,the optimization adjustment mechanism of standardized operation time,the post-standardization evaluation system and the comprehensive power outage prediction model. In-depth research of four aspects will be carried out to optimize and improve the accurate prediction ability of the comprehensive power outage operation time,and provide a reference and easy-to-operate high-efficiency auxiliary tool for the predictive deviation control of power outage operations. The practical application of production operation shows that this 10 kV distribution network power outage planning time prediction method based on big data analysis can better improve the accurate prediction ability of power outage operation time,reduce the reliability prediction deviation,and play the role of reliability target management and control to help power outage. Excellent service work such as information release is carried out accurately.

关 键 词：预测偏差 标准化 典型基准库 预测模型 

分 类 号：C39[社会学]