基于超短期负荷预测的智能配电网状态估计

Smart Power Distribution Network State Estimation Based on Extra-short Term Load Forecast

状态估计作为智能配电网自愈控制的数据出口和态势感知工具的核心板块,需要在1个数据采集周期内对全网进行1次状态估计计算,传统的配电网状态估计算法不能满足以上要求,需要研究高效的配电网状态估计算法。提出了一种基于超短期负荷预测的智能配电网的状态估计方法,为自愈控制状态评估模块和潮流计算模块提供所需数据。该算法将预测速度快、预测精度高的超短期负荷预测技术引入智能配电网状态估计,实时预测节点负荷,实现了配电系统节点负荷的实时跟踪;采用指数函数抑制不良数据的影响,提高了状态估计的精度;利用配电网潮流计算的前推回代算法计算状态变量的初始幅值和相角,提高了算法的收敛性;考虑了分布式电源接入,体现了智能电网透明开放的特点。基于IEEE36节点标准算例的计算分析,验证了算法的有效性。

As the core module of ＂data export＂ and ＂situation awareness tools＂ in smart power distribution network self-healing control, one state estimation calculation should be carried out in a data gathering period, however, the conventional state estimation calculation can＇t meet the above demands. Therefore, an efficient state estimation algorithm of smart power distribution network should be studied. Aimed to these problems, this paper presents a state estimation method for smart distribution network based on extra-short term load forecast, which could provide data to the self-healing control evaluation module and power flow calculation module. The extra-short term load forecast with fast prediction speed and high prediction accuracy is introduced to the state estimation of smart power distribution network in the algorithm, to realize the real-time prediction and track the nodal load in power distribution system. The index function is adopted to inhibit the impact of bad data, which could improve the accuracy of state estimation. In order to improve the convergence of the algorithm, forward-backward sweep method is used in the distribution network power flow calculation to calculate the initial amplitude and phase angle of state variables. The distributed power sources are permitted to access power distribution network in this algorithm, which shows the transparent and open characteristics of smart grid. Finally, the calculation and analysis based on IEEE 36 standard test example have verified the validity of the algorithm.