Monthly Water Level Prediction Based on ESMD-VMD-ESN Hybrid Model
Water level sequence contain complex features of multiple frequency information.To improve the prediction accuracy of the water level sequences,a combined model was developed based on Extreme-point Symmetric Mode Decomposition (ESMD),Variational Mode Decomposition (VMD) and Echo State Network (ESN),...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | zho |
| Published: |
Editorial Office of Pearl River
2022-01-01
|
| Series: | Renmin Zhujiang |
| Subjects: | |
| Online Access: | http://www.renminzhujiang.cn/thesisDetails?columnId=47642848&Fpath=home&index=0 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841535888365977600 |
|---|---|
| author | LI Ang ZHANG Kun SANG Yuting BI Wan |
| author_facet | LI Ang ZHANG Kun SANG Yuting BI Wan |
| author_sort | LI Ang |
| collection | DOAJ |
| description | Water level sequence contain complex features of multiple frequency information.To improve the prediction accuracy of the water level sequences,a combined model was developed based on Extreme-point Symmetric Mode Decomposition (ESMD),Variational Mode Decomposition (VMD) and Echo State Network (ESN),namely ESMD-VMD-ESN.And it was applied to forecast water level of the Taipuzha station in the upper reaches of Taipu River.The predictive effect of the “first decomposition-second decomposition-prediction-reconstruction” model was explored by comparing it with a single model ESN and the combination model ESMD-ESN.The results show that ESMD-VMD-ESN has the highest accuracy,followed by ESMD-ESN,and the lowest ESN accuracy.Compared with the ESN,the Willmott's Index of Agreement (WIA) and Pearson Correlation Coefficient (PCC) of ESMD-ESN respectively increased by 51% and 11%,the Mean Absolute Error (MAE) and Root Mean Squard Error (RMSE) of ESMD-ESN respectively decreased by 14% and 45%.ESMD can effectively simplify the water level sequence and reduce the prediction error.Compared with the ESMD-ESN,the WIA and PCC of ESMD-VMD-ESN respectively increased by 5% and 10%,the MAE and RMSE of ESMD-ESN respectively decreased by 52% and 50%.VMD can further simplify the highest frequency component of ESMD and improving the model prediction accuracy.In conclusion,the combined model ESMD-VMD-ESN has well applicability and stability in the monthly water level prediction. |
| format | Article |
| id | doaj-art-b671e6cb64c244f5a156e1c89c4a0de6 |
| institution | Kabale University |
| issn | 1001-9235 |
| language | zho |
| publishDate | 2022-01-01 |
| publisher | Editorial Office of Pearl River |
| record_format | Article |
| series | Renmin Zhujiang |
| spelling | doaj-art-b671e6cb64c244f5a156e1c89c4a0de62025-01-15T02:26:05ZzhoEditorial Office of Pearl RiverRenmin Zhujiang1001-92352022-01-014347642848Monthly Water Level Prediction Based on ESMD-VMD-ESN Hybrid ModelLI AngZHANG KunSANG YutingBI WanWater level sequence contain complex features of multiple frequency information.To improve the prediction accuracy of the water level sequences,a combined model was developed based on Extreme-point Symmetric Mode Decomposition (ESMD),Variational Mode Decomposition (VMD) and Echo State Network (ESN),namely ESMD-VMD-ESN.And it was applied to forecast water level of the Taipuzha station in the upper reaches of Taipu River.The predictive effect of the “first decomposition-second decomposition-prediction-reconstruction” model was explored by comparing it with a single model ESN and the combination model ESMD-ESN.The results show that ESMD-VMD-ESN has the highest accuracy,followed by ESMD-ESN,and the lowest ESN accuracy.Compared with the ESN,the Willmott's Index of Agreement (WIA) and Pearson Correlation Coefficient (PCC) of ESMD-ESN respectively increased by 51% and 11%,the Mean Absolute Error (MAE) and Root Mean Squard Error (RMSE) of ESMD-ESN respectively decreased by 14% and 45%.ESMD can effectively simplify the water level sequence and reduce the prediction error.Compared with the ESMD-ESN,the WIA and PCC of ESMD-VMD-ESN respectively increased by 5% and 10%,the MAE and RMSE of ESMD-ESN respectively decreased by 52% and 50%.VMD can further simplify the highest frequency component of ESMD and improving the model prediction accuracy.In conclusion,the combined model ESMD-VMD-ESN has well applicability and stability in the monthly water level prediction.http://www.renminzhujiang.cn/thesisDetails?columnId=47642848&Fpath=home&index=0Water predictionExtreme-point Symmetric Mode DecompositionVariational Mode DecompositionEcho State Network |
| spellingShingle | LI Ang ZHANG Kun SANG Yuting BI Wan Monthly Water Level Prediction Based on ESMD-VMD-ESN Hybrid Model Renmin Zhujiang Water prediction Extreme-point Symmetric Mode Decomposition Variational Mode Decomposition Echo State Network |
| title | Monthly Water Level Prediction Based on ESMD-VMD-ESN Hybrid Model |
| title_full | Monthly Water Level Prediction Based on ESMD-VMD-ESN Hybrid Model |
| title_fullStr | Monthly Water Level Prediction Based on ESMD-VMD-ESN Hybrid Model |
| title_full_unstemmed | Monthly Water Level Prediction Based on ESMD-VMD-ESN Hybrid Model |
| title_short | Monthly Water Level Prediction Based on ESMD-VMD-ESN Hybrid Model |
| title_sort | monthly water level prediction based on esmd vmd esn hybrid model |
| topic | Water prediction Extreme-point Symmetric Mode Decomposition Variational Mode Decomposition Echo State Network |
| url | http://www.renminzhujiang.cn/thesisDetails?columnId=47642848&Fpath=home&index=0 |
| work_keys_str_mv | AT liang monthlywaterlevelpredictionbasedonesmdvmdesnhybridmodel AT zhangkun monthlywaterlevelpredictionbasedonesmdvmdesnhybridmodel AT sangyuting monthlywaterlevelpredictionbasedonesmdvmdesnhybridmodel AT biwan monthlywaterlevelpredictionbasedonesmdvmdesnhybridmodel |