基于数据驱动的配电网线损计算方法研究  

Research on Data-Driven-Based Line Loss Calculation Method for Distribution Network

作　　者：刘亦驰 刘建平 李雄 LIU Yichi;LIU Jianping;LI Xiong(Information Center,Guizhou Power Grid Co.,Ltd.,Guiyang 550003,China;Guizhou Power Grid Co.,Ltd.,Guiyang 550002,China)

机构地区：[1]贵州电网有限责任公司信息中心,贵州贵阳550003 [2]贵州电网有限责任公司,贵州贵阳550002

出　　处：《自动化仪表》2025年第3期111-115,121,共6页Process Automation Instrumentation

摘　　要：针对目前配电网线损计算时获取数据过程中异常测量导致计算误差较大的问题,提出了一种基于数据驱动的配电网线损计算方法。首先,考虑到分布式电源配电网潮流的变化特性,设置多个变量来表示一段时间内不同时间点的潮流状态,从而提高线损计算的准确性。其次,建立了一个非线性优化模型。该模型包含正常数据采集总线的功率约束和异常或缺失数据总线的上下功率约束。最后,建立以配电网第一支路传输功率的上限和下限为约束的线性模型,从而提高区间估计的可靠性。试验结果表明,实际线损在线损区间值内,且最大误差不超过0.13 kW。所提方法对提高配电网线损计算准确性具有一定借鉴作用。Aiming at the current problem of large calculation errors caused by abnormal measurements in the process of acquiring data for distribution network line loss calculation,a data-driven-based line loss calculation method for distribution network is proposed.Firstly,considering the changing characteristics of the distributed power distribution network currents,multiple variables are set to represent the current state at different time points over a period,so as to improve the accuracy of line loss calculation.Secondly,a nonlinear optimization model is established.The model contains power constraints for normal data collection buses and upper and lower power constraints for abnormal or missing data buses.Finally,a linear model with the upper and lower constraints of the transmission power of the first branch of the distribution network is established to improve the reliability of interval estimation.The experimental results show that the actual line loss is within the line loss interval value and the maximum error is not more than 0.13 kW.The proposed method is useful for improving the accuracy of line loss calculation in distribution networks.

关 键 词：配电网 数据驱动 线损计算 分布式电源 区间模型 潮流方程 功率约束 非线性优化 

分 类 号：TH-39[机械工程]