摘要
针对配电网状态估计实时量测数量的不足,提出了一种基于ANN伪量测建模的配电网状态估计算法。该方法采用人工神经网络网络(ANN),将部分实时量测数据作为神经网络的输入,产生较为精确的负荷伪量测数据。此外,应用高斯混合模型对产生伪量测的误差进行分解拟合,从而获得负荷伪量测的权重。最后,将获得的伪量测及其权重输入到状态估计模块中,实现了配电网的状态估计。通过英国标准配网系统(UKGDS)中16节点模型的仿真结果表明,该算法提高了配电网状态估计的精度,具有一定的现实意义和理论价值。
Considering the lack of real-time measurements, an Artificial Neural Network(ANN)for pseudo measurement modeling based state estimation algorithm is proposed for distribution networks. In the proposed approach, pseudo measurements are generated from a few real measurements using artificial neural networks. In addition, the Gaussian mixture model is used to decompose errors of pseudo measurements and produce its weight. This algorithm considerably improves the accuracy of computation. Test results for UKGDS 16-bus test system are presented to show the value of the algorithm proposed on both practice and theory.
作者
张明光
张钰
ZHANG Mingguang;ZHANG Yu(School of Electrical and Information Engineering, Lanzhou University of Technology, Lanzhou 730050, China)
出处
《计算机工程与应用》
CSCD
北大核心
2016年第17期253-256,260,共5页
Computer Engineering and Applications
基金
国家自然科学基金(No.51467009)
关键词
配电网
状态估计
人工神经网络
高斯混合模型
伪量测
distribution network
state estimation
artificial neural network
Gaussian mixture model
pseudo measurement
