基于改进神经网络的电网线损率预测模型分析  

Analysis of Line Loss Rate Prediction Model of Power Grid Based on Improved Neural Network

作　　者：陈广宇[1] 张磊 赵磊 张慧敏 李磐旎 CHEN Guang-yu;ZHANG Lei;ZHAO Lei;ZHANG Hui-min;LI Pan-ni(Chengde Power Supply Company,State Grid Jibei Electric Power Company Limited,Chengde 067000 China;Beijing Join Bright Digital Power Technology Co.,Ltd.,Beijing 100085 China)

机构地区：[1]国网冀北电力有限公司承德供电公司,河北承德067000 [2]北京中恒博瑞数字电力科技有限公司,北京100085

出　　处：《自动化技术与应用》2022年第5期130-133,共4页Techniques of Automation and Applications

基　　金：国网冀北电力有限公司科技项目(52018F18000T)。

摘　　要：对神经网络输入层中的数据作归一化处理,通过优化神经网络可有效避免神经元出现饱和现象,从而提高对线损预测的准确性。因此,本研究利用改进神经网络设计了新的电网线损率预测模型。实验结果表明,该预测模型的预测时效性和预测准确率均较高,证明该模型具有较强的应用优势。By normalizing the data in the input layer of the neural network,the saturation of neurons can be effectively avoided by optimizing the neural network,so as to improve the accuracy of line loss prediction.Therefore,this study uses the improved neural network to design a new network line loss rate prediction model.The experimental results show that the prediction model has high prediction timeliness and accuracy,which proves that the model has strong application advantages.

关 键 词：改进神经网络 线损率预测 等值电阻法 归一化 灰色关联分析 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程]