A new depth model of domain fusion (DWT-CNN-LSTM) was proposed to accurately complete the prediction based on discrete wavelet transform (DWT), convolutional neural network (CNN) and long short-term memory (LSTM). Two channels of the model were proposed: the original channel ...
同等输入条件下,Prophet耦合模型较LSTM耦合模型预测表现效果更佳.(4)针对预测模型径流输入的非线性特征,采用CEEMD和DWT两种分解方法,将原始径流序列分解成多个子序列... 许晶晶 - 长安大学 被引量: 0发表: 2023年 基于DWT-CNN-LSTM 的超短期光伏发电功率预测 To make photovoltaic power generation stable, the inf...
network,CNN)和长短期记忆神经网络(long short-term memory,LSTM)的新型域融合深度模型(DWT-CNN-LSTM),以准确地完成预测.提出的模型具有两个通道:原始通道和DWT通道.CNN分别从原始通道和DWT通道提取时域和频域特征,LSTM则用以挖掘具有长期依赖性的特征,从而形成具有长期依赖性的时域和频域的融合特征,可用于功率预测....