Research and Application of Data-driven Forecast Method for Water Level Increment of Small and Medium-sized Rivers
Focusing on the forecast of flash flood disasters in small and medium-sized rivers,this paper adopts three data-driven forecast models to forecast the maximum water level increment.With Paitan Town,Zengcheng District,Guangzhou City as the research area,the mathematical modeling and accuracy evaluati...
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.1001-9235.2023.08.011 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841536142102495232 |
|---|---|
| author | DING Wu LIN Hanxiong WANG Hangang ZHANG Wei YANG Bin |
| author_facet | DING Wu LIN Hanxiong WANG Hangang ZHANG Wei YANG Bin |
| author_sort | DING Wu |
| collection | DOAJ |
| description | Focusing on the forecast of flash flood disasters in small and medium-sized rivers,this paper adopts three data-driven forecast models to forecast the maximum water level increment.With Paitan Town,Zengcheng District,Guangzhou City as the research area,the mathematical modeling and accuracy evaluation of the three models are carried out.The analysis shows that the forecast accuracy of machine learning is the highest,but the forecast results are not interpretable.The forecast accuracy of the similarity analysis model is medium,and the prediction results have good cognition and interpretability.The overall prediction accuracy of the “rainfall-water level increment” relationship model is the lowest,but its operation is simple.The calculation of the similarity analysis model and machine learning model is relatively complex,relying on computer-aided calculation,and the results of the “rainfall-water level increment” relationship model can be plotted as graphs.Considering the forecast accuracy and convenience,multiple models such as the machine learning model and similarity analysis model can be applied for the forecast to improve the forecast accuracy if there is support from auxiliary equipment such as computers and mobile phone applications.Without support from computers and other basic equipment,the “rainfall-water level increment” relationship model can provide a convenient forecast method for flash flood forecast and early warning in small and medium-sized rivers. |
| format | Article |
| id | doaj-art-e8fabfaa3e084bf688a727071c31b42d |
| institution | Kabale University |
| issn | 1001-9235 |
| language | zho |
| publishDate | 2023-01-01 |
| publisher | Editorial Office of Pearl River |
| record_format | Article |
| series | Renmin Zhujiang |
| spelling | doaj-art-e8fabfaa3e084bf688a727071c31b42d2025-01-15T02:21:49ZzhoEditorial Office of Pearl RiverRenmin Zhujiang1001-92352023-01-014447637194Research and Application of Data-driven Forecast Method for Water Level Increment of Small and Medium-sized RiversDING WuLIN HanxiongWANG HangangZHANG WeiYANG BinFocusing on the forecast of flash flood disasters in small and medium-sized rivers,this paper adopts three data-driven forecast models to forecast the maximum water level increment.With Paitan Town,Zengcheng District,Guangzhou City as the research area,the mathematical modeling and accuracy evaluation of the three models are carried out.The analysis shows that the forecast accuracy of machine learning is the highest,but the forecast results are not interpretable.The forecast accuracy of the similarity analysis model is medium,and the prediction results have good cognition and interpretability.The overall prediction accuracy of the “rainfall-water level increment” relationship model is the lowest,but its operation is simple.The calculation of the similarity analysis model and machine learning model is relatively complex,relying on computer-aided calculation,and the results of the “rainfall-water level increment” relationship model can be plotted as graphs.Considering the forecast accuracy and convenience,multiple models such as the machine learning model and similarity analysis model can be applied for the forecast to improve the forecast accuracy if there is support from auxiliary equipment such as computers and mobile phone applications.Without support from computers and other basic equipment,the “rainfall-water level increment” relationship model can provide a convenient forecast method for flash flood forecast and early warning in small and medium-sized rivers.http://www.renminzhujiang.cn/thesisDetails#10.3969/j.issn.1001-9235.2023.08.011data-drivenmachine learningwater level increment forecastflash flood early warning |
| spellingShingle | DING Wu LIN Hanxiong WANG Hangang ZHANG Wei YANG Bin Research and Application of Data-driven Forecast Method for Water Level Increment of Small and Medium-sized Rivers Renmin Zhujiang data-driven machine learning water level increment forecast flash flood early warning |
| title | Research and Application of Data-driven Forecast Method for Water Level Increment of Small and Medium-sized Rivers |
| title_full | Research and Application of Data-driven Forecast Method for Water Level Increment of Small and Medium-sized Rivers |
| title_fullStr | Research and Application of Data-driven Forecast Method for Water Level Increment of Small and Medium-sized Rivers |
| title_full_unstemmed | Research and Application of Data-driven Forecast Method for Water Level Increment of Small and Medium-sized Rivers |
| title_short | Research and Application of Data-driven Forecast Method for Water Level Increment of Small and Medium-sized Rivers |
| title_sort | research and application of data driven forecast method for water level increment of small and medium sized rivers |
| topic | data-driven machine learning water level increment forecast flash flood early warning |
| url | http://www.renminzhujiang.cn/thesisDetails#10.3969/j.issn.1001-9235.2023.08.011 |
| work_keys_str_mv | AT dingwu researchandapplicationofdatadrivenforecastmethodforwaterlevelincrementofsmallandmediumsizedrivers AT linhanxiong researchandapplicationofdatadrivenforecastmethodforwaterlevelincrementofsmallandmediumsizedrivers AT wanghangang researchandapplicationofdatadrivenforecastmethodforwaterlevelincrementofsmallandmediumsizedrivers AT zhangwei researchandapplicationofdatadrivenforecastmethodforwaterlevelincrementofsmallandmediumsizedrivers AT yangbin researchandapplicationofdatadrivenforecastmethodforwaterlevelincrementofsmallandmediumsizedrivers |