基于ResNet+GRU组合模型的负电荷预测方法研究  

Research on Negative Charge Prediction Method Based on ResNet+GRU Combination Model

作　　者：刘义卿 陈新房[1] 赵晗清 LIU Yi-qing;CHEN Xin-fang;ZHAO Han-qing(Institute of Disaster Prevention,Langfang 065201,Hebei)

机构地区：[1]防灾科技学院,河北廊坊065201

出　　处：《电脑与电信》2023年第5期89-94,共6页Computer & Telecommunication

基　　金：2022防灾科技学院大学生创新创业项目,项目编号:S202211775045。

摘　　要：电力负荷预测在保证电力系统安全运行、平衡能源分配等方面至关重要,精准的负荷预测能有效减少用电事故的发生,提高用电效率。利用神经网络组合模型的优势,充分提取时间序列的特征,输入到GRU网络进行预测,使得模型的预测准确率更高、泛化能力更强。使用ResNet解决卷积网络层数堆叠带来的问题,将处理后的数据作为GRU的输入,以此构建泛化能力更强的混合模型ResNet+GRU。为了验证所提出的技术,本实验选取100个数据节点,并通过实验证明ResNet+GRU优于所有方法,使用ResNet+GRU获得MAE、MSE、RMSE和MAPE的值最小。Power load forecasting is crucial in ensuring the safe operation of the power system and balancing energy distribution.Accurate load forecasting can effectively reduce the occurrence of electricity accidents and improve electricity efficiency.By utilizing the advantages of neural network combination models,the features of time series are fully extracted and input into the GRU network for prediction,resulting in higher prediction accuracy and stronger generalization ability of the model.Using ResNet to solve the problem of convolutional network layer stacking,the processed data is used as the input of GRU to construct a hybrid model ResNet+GRU with stronger generalization ability.To validate the proposed technique,100 data nodes are selected in this experiment,and it is demonstrated that ResNet+GRU is superior to all methods.ResNet+GRU is used to obtain the minimum values of MAE,MSE,RMSE,and MAPE.

关 键 词：电力负荷预测 ResNet GRU ResNet+GRU 

分 类 号：TM715[电气工程—电力系统及自动化] TP183[自动化与计算机技术—控制理论与控制工程]