基于迁移学习的CNN-GRU短期电力负荷预测方法

Transfer learning based CNN‐GRU short‐term power load forecasting method

随着信息化时代的到来,数据驱动的机器学习已成为主流的短期电力负荷预测方法,但此类方法对数据具有很强的依赖性。为改善这一不足,提出了一种基于迁移学习的卷积神经网络(convolutional neural network,CNN)和门控循环单元(gated recurrent unit,GRU)的混合预测模型。以东北某2个地区的电力负荷数据集为例进行预测分析。结果表明,所提方法可有效提高数据稀缺情况下短期电力负荷预测的准确性和可靠性。

With the advent of the information era,data-driven machine learning has become a mainstream shortterm power load forecasting method,but such method is highly dependent on data.In order to improve this deficiency,this paper proposes a hybrid prediction model based on convolutional neural network(CNN)and gated recurrent unit(GRU)of transfer learning.Two power load data sets A and B in northeast China are taken as examples to forecast and analyze the short-term power load,the results show that the proposed method can effectively improve the accuracy and reliability of short-term power load forecasting in the case of data scarcity.