基于神经网络模型的智能电网线损估计方法  

Estimation method for line loss of smart grid basedon neural network model

作　　者：李果 袁小凯 黄世平 LI Guo;YUAN Xiao-kai;HUANG Shi-ping(Research Institute,China Southern Power Grid Co.Ltd.,Guangzhou 510080,China)

机构地区：[1]中国南方电网有限公司科学研究院,广州510080

出　　处：《沈阳工业大学学报》2022年第2期133-138,共6页Journal of Shenyang University of Technology

基　　金：国家自然科学基金青年科学基金项目(61101249);中国南方电网数字化转型项目(SEPRI-H182022).

摘　　要：为了准确、有效、实时估计智能电网中配电网线损,提出了一种基于神经网络模型的智能电网线损估计方法.在BP神经网络算法的基础上采用LM算法对神经网络权重和阈值进行连续优化从而实现网络自适应调节,进而搭建神经网络模型.将模型应用于IEEE33节点系统进行实验,实时估计每条线路的功率损耗并将估计线损值与实际测得的线损值进行比较并提出相应的评估指标对方法有效性进行评估.结果表明,与传统的潮流法相比,所提出的方法具有更优的运算速度和准确度.For the accurate,effective and real-time estimation of line loss of distribution network in the smart grid,an estimation method for the line loss of smart grid based on neural network model was proposed.Based on a BP neural network algorithm,the weights and thresholds of neural network were continuously optimized with a LM algorithm to achieve adaptive network adjustment,and then to build the neural network model.The model was applied to an IEEE33 node system for experiments,the power loss of each line was estimated in real time and compared with the line loss values experimentally measured,and corresponding evaluation indexes were proposed to evaluate the effectiveness of method.The results show that the as-proposed method is better owing to its calculation speed and accuracy compared with the traditional power flow methods.

关 键 词：神经网络模型 BP神经网络算法 LM算法 线路损耗 实时估计 智能电网 IEEE33节点系统 配电网 潮流法 

分 类 号：TM727[电气工程—电力系统及自动化]