摘要
针对现有方法难以快速准确得到中低压配电系统网损的问题,考虑以数据关联关系替代复杂物理模型,提出一种三相不平衡配电系统网损的快速估计方法。先定义了表征系统三相负荷不平衡水平的负荷不平衡系数,并结合网损理论计算来模拟生成系统网损样本数据;再定性分析负荷水平和不平衡水平与系统总网损间关联性,构建深度信念网络(DBN)学习样本数据,获得特征量间关联关系的隐式表征。算例仿真表明,训练后的DBN能够有效表征负荷不平衡系数和负荷视在功率与系统总网损间的复杂映射关系,可快速准确估计系统总网损。
It is difficult for the existing methods to quickly and accurately obtain the power loss of MV and LV distribution system.A fast estimation method,in which the data correlation relationship is considered to replace the complex physical model,for the power loss of three-phase unbalanced distribution system is proposed in this paper.We define the load unbalance index to characterize the three-phase load unbalance level,and the network loss theoretical calculation is used to simulate the generation of system network loss sample data.Then we qualitatively analyze the relationship between the load level and the unbalance level and the total system network loss,and the deep belief network(DBN) is constructed to learn the implicit representations of the correlations between features.The simulations show that the trained DBN can effectively represent the complexity mapping between the load unbalance index and the apparent load power with the total loss,and can quickly and accurately estimate the total power loss.
作者
王瑞
赵莉华
胡帅
WANG Rui;ZHAO Li-hua;HU Shuai(College of Electrical Engineering,Sichuan University,Chengdu 610065,China;State Grid Ningxia Electric Power Eco-Tech Research Institute,Yinchuan 750004,China)
出处
《水电能源科学》
北大核心
2021年第5期202-206,共5页
Water Resources and Power
基金
中央高校基本科研业务费专项资金项目(YJ201655)。
关键词
配电系统
三相不平衡
网损
负荷不平衡系数
深度信念网络
distribution network
three-phase unbalance
power loss
load unbalance factor
deep belief network