Water level prediction of Liuxihe Reservoir based on improved long short-term memory neural network
To meet the demand of accurate water level prediction of the reservoir in Liuxihe River Basin, this paper proposes an improved long short-term memory (LSTM) neural network based on the Bayesian optimization algorithm and wavelet decomposition coupling. Based on the improved model, the water levels o...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IWA Publishing
2023-11-01
|
| Series: | Water Supply |
| Subjects: | |
| Online Access: | http://ws.iwaponline.com/content/23/11/4563 |
| _version_ | 1797428376037752832 |
|---|---|
| author | Youming Li Jia Qu Haosen Zhang Yan Long Shu Li |
| author_facet | Youming Li Jia Qu Haosen Zhang Yan Long Shu Li |
| author_sort | Youming Li |
| collection | DOAJ |
| description | To meet the demand of accurate water level prediction of the reservoir in Liuxihe River Basin, this paper proposes an improved long short-term memory (LSTM) neural network based on the Bayesian optimization algorithm and wavelet decomposition coupling. Based on the improved model, the water levels of Liuxihe Reservoir and Huanglongdai Reservoir are simulated and predicted by the 1 h prediction length, and the prediction accuracy of the improved model is verified separately by the 3, 6 and 12 h prediction lengths. The results show that: first, Bayesian optimization coupling can significantly reduce the average absolute error and root mean square error of the model and improve the overall prediction accuracy, but this algorithm is insufficient in the optimization of model extremum; Wavelet decomposition coupling can significantly reduce the outliers in model prediction and improve the accuracy of extremum, but it plays relatively weaker role in the overall optimization of the model. Second, by the prediction lengths of 1, 3, 6 and 12 h, the improved model based on the LSTM neural network and coupled with Bayesian optimization and wavelet decomposition is superior to Bayesian optimization and wavelet decomposition coupling model in overall prediction accuracy and prediction accuracy of extremum.
HIGHLIGHTS
Quantifying the impacts of coupling model on reservoir water level.;
Proposing an improved prediction model for long short-term memory neural network.;
Quantifying the impacts of different input characteristics on the model.; |
| first_indexed | 2024-03-09T08:58:00Z |
| format | Article |
| id | doaj.art-495a0f969c2d40a1b09f647006d7480a |
| institution | Directory Open Access Journal |
| issn | 1606-9749 1607-0798 |
| language | English |
| last_indexed | 2024-03-09T08:58:00Z |
| publishDate | 2023-11-01 |
| publisher | IWA Publishing |
| record_format | Article |
| series | Water Supply |
| spelling | doaj.art-495a0f969c2d40a1b09f647006d7480a2023-12-02T12:34:12ZengIWA PublishingWater Supply1606-97491607-07982023-11-0123114563458210.2166/ws.2023.282282Water level prediction of Liuxihe Reservoir based on improved long short-term memory neural networkYouming Li0Jia Qu1Haosen Zhang2Yan Long3Shu Li4 School of Water Resources and Electric Power, Hebei University of Engineering, Handan, Hebei Province, China School of Water Resources and Electric Power, Hebei University of Engineering, Handan, Hebei Province, China School of Water Resources and Electric Power, Hebei University of Engineering, Handan, Hebei Province, China School of Water Resources and Electric Power, Hebei University of Engineering, Handan, Hebei Province, China Tongliao Water Conservancy Development Center, Inner Mongolia, China To meet the demand of accurate water level prediction of the reservoir in Liuxihe River Basin, this paper proposes an improved long short-term memory (LSTM) neural network based on the Bayesian optimization algorithm and wavelet decomposition coupling. Based on the improved model, the water levels of Liuxihe Reservoir and Huanglongdai Reservoir are simulated and predicted by the 1 h prediction length, and the prediction accuracy of the improved model is verified separately by the 3, 6 and 12 h prediction lengths. The results show that: first, Bayesian optimization coupling can significantly reduce the average absolute error and root mean square error of the model and improve the overall prediction accuracy, but this algorithm is insufficient in the optimization of model extremum; Wavelet decomposition coupling can significantly reduce the outliers in model prediction and improve the accuracy of extremum, but it plays relatively weaker role in the overall optimization of the model. Second, by the prediction lengths of 1, 3, 6 and 12 h, the improved model based on the LSTM neural network and coupled with Bayesian optimization and wavelet decomposition is superior to Bayesian optimization and wavelet decomposition coupling model in overall prediction accuracy and prediction accuracy of extremum. HIGHLIGHTS Quantifying the impacts of coupling model on reservoir water level.; Proposing an improved prediction model for long short-term memory neural network.; Quantifying the impacts of different input characteristics on the model.;http://ws.iwaponline.com/content/23/11/4563deep learningforesight periodlong short-term memoryreservoir levelwater level prediction |
| spellingShingle | Youming Li Jia Qu Haosen Zhang Yan Long Shu Li Water level prediction of Liuxihe Reservoir based on improved long short-term memory neural network Water Supply deep learning foresight period long short-term memory reservoir level water level prediction |
| title | Water level prediction of Liuxihe Reservoir based on improved long short-term memory neural network |
| title_full | Water level prediction of Liuxihe Reservoir based on improved long short-term memory neural network |
| title_fullStr | Water level prediction of Liuxihe Reservoir based on improved long short-term memory neural network |
| title_full_unstemmed | Water level prediction of Liuxihe Reservoir based on improved long short-term memory neural network |
| title_short | Water level prediction of Liuxihe Reservoir based on improved long short-term memory neural network |
| title_sort | water level prediction of liuxihe reservoir based on improved long short term memory neural network |
| topic | deep learning foresight period long short-term memory reservoir level water level prediction |
| url | http://ws.iwaponline.com/content/23/11/4563 |
| work_keys_str_mv | AT youmingli waterlevelpredictionofliuxihereservoirbasedonimprovedlongshorttermmemoryneuralnetwork AT jiaqu waterlevelpredictionofliuxihereservoirbasedonimprovedlongshorttermmemoryneuralnetwork AT haosenzhang waterlevelpredictionofliuxihereservoirbasedonimprovedlongshorttermmemoryneuralnetwork AT yanlong waterlevelpredictionofliuxihereservoirbasedonimprovedlongshorttermmemoryneuralnetwork AT shuli waterlevelpredictionofliuxihereservoirbasedonimprovedlongshorttermmemoryneuralnetwork |