基于随机森林算法的台区合理线损率估计方法

发布时间：2018-03-22 05:37

本文选题：台区线损率　切入点：线损管理　出处：《电力自动化设备》2017年11期 　论文类型：期刊论文

【摘要】：线损管理是电力公司的重点管理内容之一,低压电网普遍采用分台区的管理手段。供电侧数据缺失和营销抄表日期冲突导致的线损率缺失是电力公司线损系统中台区线损数据存在的主要问题。为此,提出了一种涉及多源数据的基于随机森林算法的台区合理线损率估计方法。从线损系统、生产管理系统和营销系统中提取台区、变压器和用户相关数据,建立台区特征数据库;对台区进行聚类分析,并在此基础上建立决策树分类模型和随机森林估计模型;利用上述模型估计台区线损率。以上海电力公司实际数据为例,计算结果验证了所提方法的可行性;并将所得结果与线性回归模型和回归树模型的估计结果进行比较,表明所提方法性能优越。
[Abstract]:Line loss management is one of the key contents of the management of electric power company, low-voltage power grid is widely used in Taiwan area. The line loss management means the supply side data missing and marketing meter reading date conflict rate is the lack of main problems of line loss system in Taiwan District line loss data. Therefore, proposed a multi-source data involving random forest algorithm based on the Taiwan area reasonable loss rate estimation method. The extraction of line loss system, Taiwan production management system and marketing system, the related data of transformer and users, establish the Taiwan area feature database; cluster analysis of Taiwan area, based on the decision tree classification model and random forest model estimation; estimation of line loss rate by the model with the actual data of Shanghai Power Co. For example, the calculation results verify the feasibility of the proposed method; and the results obtained from the linear regression model and regression The comparison of the estimation results of the tree model shows that the proposed method is superior in performance.

【作者单位】：天津大学智能电网教育部重点实验室;国网上海市电力公司;
【基金】：国家高技术研究发展计划(863计划)资助项目(2015-AA050203)~~
【分类号】：TM73;TM744

【相似文献】