Ensemble streamflow forecasting based on variational mode decomposition and long short term memory
Abstract Reliable and accurate streamflow forecasting plays a vital role in the optimal management of water resources. To improve the stability and accuracy of streamflow forecasting, a hybrid decomposition-ensemble model named VMD-LSTM-GBRT, which is sensitive to sampling, noise and long historical...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2022-01-01
|
| Series: | Scientific Reports |
| Online Access: | https://doi.org/10.1038/s41598-021-03725-7 |
| _version_ | 1798026506114433024 |
|---|---|
| author | Xiaomei Sun Haiou Zhang Jian Wang Chendi Shi Dongwen Hua Juan Li |
| author_facet | Xiaomei Sun Haiou Zhang Jian Wang Chendi Shi Dongwen Hua Juan Li |
| author_sort | Xiaomei Sun |
| collection | DOAJ |
| description | Abstract Reliable and accurate streamflow forecasting plays a vital role in the optimal management of water resources. To improve the stability and accuracy of streamflow forecasting, a hybrid decomposition-ensemble model named VMD-LSTM-GBRT, which is sensitive to sampling, noise and long historical changes of streamflow, was established. The variational mode decomposition (VMD) algorithm was first applied to extract features, which were then learned by several long short-term memory (LSTM) networks. Simultaneously, an ensemble tree, a gradient boosting tree for regression (GBRT), was trained to model the relationships between the extracted features and the original streamflow. The outputs of these LSTMs were finally reconstructed by the GBRT model to obtain the forecasting streamflow results. A historical daily streamflow series (from 1/1/1997 to 31/12/2014) for Yangxian station, Han River, China, was investigated by the proposed model. VMD-LSTM-GBRT was compared with respect to three aspects: (1) feature extraction algorithm; ensemble empirical mode decomposition (EEMD) was used. (2) Feature learning techniques; deep neural networks (DNNs) and support vector machines for regression (SVRs) were exploited. (3) Ensemble strategy; the summation strategy was used. The results indicate that the VMD-LSTM-GBRT model overwhelms all other peer models in terms of the root mean square error (RMSE = 36.3692), determination coefficient (R2 = 0.9890), mean absolute error (MAE = 9.5246) and peak percentage threshold statistics (PPTS(5) = 0.0391%). The addressed approach based on the memory of long historical changes with deep feature representations had good stability and high prediction precision. |
| first_indexed | 2024-04-11T18:36:31Z |
| format | Article |
| id | doaj.art-81c254d86c4943889083a1e196592bd6 |
| institution | Directory Open Access Journal |
| issn | 2045-2322 |
| language | English |
| last_indexed | 2024-04-11T18:36:31Z |
| publishDate | 2022-01-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Scientific Reports |
| spelling | doaj.art-81c254d86c4943889083a1e196592bd62022-12-22T04:09:14ZengNature PortfolioScientific Reports2045-23222022-01-0112111910.1038/s41598-021-03725-7Ensemble streamflow forecasting based on variational mode decomposition and long short term memoryXiaomei Sun0Haiou Zhang1Jian Wang2Chendi Shi3Dongwen Hua4Juan Li5Shaanxi Provincial Land Engineering Construction Group Co., Ltd.Shaanxi Provincial Land Engineering Construction Group Co., Ltd.Shaanxi Provincial Land Engineering Construction Group Co., Ltd.Shaanxi Provincial Land Engineering Construction Group Co., Ltd.Shaanxi Provincial Land Engineering Construction Group Co., Ltd.Shaanxi Provincial Land Engineering Construction Group Co., Ltd.Abstract Reliable and accurate streamflow forecasting plays a vital role in the optimal management of water resources. To improve the stability and accuracy of streamflow forecasting, a hybrid decomposition-ensemble model named VMD-LSTM-GBRT, which is sensitive to sampling, noise and long historical changes of streamflow, was established. The variational mode decomposition (VMD) algorithm was first applied to extract features, which were then learned by several long short-term memory (LSTM) networks. Simultaneously, an ensemble tree, a gradient boosting tree for regression (GBRT), was trained to model the relationships between the extracted features and the original streamflow. The outputs of these LSTMs were finally reconstructed by the GBRT model to obtain the forecasting streamflow results. A historical daily streamflow series (from 1/1/1997 to 31/12/2014) for Yangxian station, Han River, China, was investigated by the proposed model. VMD-LSTM-GBRT was compared with respect to three aspects: (1) feature extraction algorithm; ensemble empirical mode decomposition (EEMD) was used. (2) Feature learning techniques; deep neural networks (DNNs) and support vector machines for regression (SVRs) were exploited. (3) Ensemble strategy; the summation strategy was used. The results indicate that the VMD-LSTM-GBRT model overwhelms all other peer models in terms of the root mean square error (RMSE = 36.3692), determination coefficient (R2 = 0.9890), mean absolute error (MAE = 9.5246) and peak percentage threshold statistics (PPTS(5) = 0.0391%). The addressed approach based on the memory of long historical changes with deep feature representations had good stability and high prediction precision.https://doi.org/10.1038/s41598-021-03725-7 |
| spellingShingle | Xiaomei Sun Haiou Zhang Jian Wang Chendi Shi Dongwen Hua Juan Li Ensemble streamflow forecasting based on variational mode decomposition and long short term memory Scientific Reports |
| title | Ensemble streamflow forecasting based on variational mode decomposition and long short term memory |
| title_full | Ensemble streamflow forecasting based on variational mode decomposition and long short term memory |
| title_fullStr | Ensemble streamflow forecasting based on variational mode decomposition and long short term memory |
| title_full_unstemmed | Ensemble streamflow forecasting based on variational mode decomposition and long short term memory |
| title_short | Ensemble streamflow forecasting based on variational mode decomposition and long short term memory |
| title_sort | ensemble streamflow forecasting based on variational mode decomposition and long short term memory |
| url | https://doi.org/10.1038/s41598-021-03725-7 |
| work_keys_str_mv | AT xiaomeisun ensemblestreamflowforecastingbasedonvariationalmodedecompositionandlongshorttermmemory AT haiouzhang ensemblestreamflowforecastingbasedonvariationalmodedecompositionandlongshorttermmemory AT jianwang ensemblestreamflowforecastingbasedonvariationalmodedecompositionandlongshorttermmemory AT chendishi ensemblestreamflowforecastingbasedonvariationalmodedecompositionandlongshorttermmemory AT dongwenhua ensemblestreamflowforecastingbasedonvariationalmodedecompositionandlongshorttermmemory AT juanli ensemblestreamflowforecastingbasedonvariationalmodedecompositionandlongshorttermmemory |