Runoff Prediction Based on VMD-LSTM-ARMA Model
To increase the accuracy of runoff prediction,a combined prediction model is proposed to solve the problems of runoff prediction.This model consists of the variational mode decomposition (VMD),long and short-term memory (LSTM) network,and autoregressive moving average (ARMA).To reduce the complexity...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | zho |
| Published: |
Editorial Office of Pearl River
2023-01-01
|
| Series: | Renmin Zhujiang |
| Subjects: | |
| Online Access: | http://www.renminzhujiang.cn/thesisDetails#10.3969/j.issn.1001-9235.2023.04.012 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | To increase the accuracy of runoff prediction,a combined prediction model is proposed to solve the problems of runoff prediction.This model consists of the variational mode decomposition (VMD),long and short-term memory (LSTM) network,and autoregressive moving average (ARMA).To reduce the complexity of the incoming traffic,the runoff data is decomposed into three modal components with different frequencies by VMD.The low-frequency modal component inherits the temporal properties of the data and can be processed by constructing the LSTM prediction model,while the two high-frequency sequences are stationary time series and can be processed by constructing the ARMA prediction model.The prediction results of the three subsequences are superimposed to predict the runoff.Finally,the hourly flow data of Dongjiang Hydrometric Station of Xiangjiang river tributary in 2020 are used to forecast the inflow flow,and the results of comparative experiments with other algorithms show that the constructed model can effectively improve the accuracy of the hydrological forecast. |
|---|---|
| ISSN: | 1001-9235 |