Enhancing a Multi-Step Discharge Prediction with Deep Learning and a Response Time Parameter
Flood forecasting is among the most important precaution measures to prevent devastating disasters affecting human life, properties, and the overall environment. It is closely involved with precipitation and streamflow data forecasting tasks. In this work, we introduced a multi-step discharge predic...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-09-01
|
| Series: | Water |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2073-4441/14/18/2898 |
| _version_ | 1797481202194579456 |
|---|---|
| author | Wandee Thaisiam Warintra Saelo Papis Wongchaisuwat |
| author_facet | Wandee Thaisiam Warintra Saelo Papis Wongchaisuwat |
| author_sort | Wandee Thaisiam |
| collection | DOAJ |
| description | Flood forecasting is among the most important precaution measures to prevent devastating disasters affecting human life, properties, and the overall environment. It is closely involved with precipitation and streamflow data forecasting tasks. In this work, we introduced a multi-step discharge prediction framework based on deep learning models. A simple feature representation technique using a correlation of backward lags was enhanced with a time of concentration (T<sub>C</sub>) concept. Recurrent neural networks and their variants, coupled with the T<sub>C</sub>-related features, provided superior performance with over 0.9 Nash–Sutcliffe model efficiency coefficient and substantially high correlation values for multiple forecasted points. These results were consistent among both the Upper Nan and the Loei river basins in Thailand, which were used as case studies in this work. |
| first_indexed | 2024-03-09T22:11:08Z |
| format | Article |
| id | doaj.art-2eb0717650ca4b2fa6f646e0756b8d89 |
| institution | Directory Open Access Journal |
| issn | 2073-4441 |
| language | English |
| last_indexed | 2024-03-09T22:11:08Z |
| publishDate | 2022-09-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Water |
| spelling | doaj.art-2eb0717650ca4b2fa6f646e0756b8d892023-11-23T19:31:54ZengMDPI AGWater2073-44412022-09-011418289810.3390/w14182898Enhancing a Multi-Step Discharge Prediction with Deep Learning and a Response Time ParameterWandee Thaisiam0Warintra Saelo1Papis Wongchaisuwat2Department of Water Resources Engineering, Kasetsart University, Bangkok 10900, ThailandDepartment of Water Resources Engineering, Kasetsart University, Bangkok 10900, ThailandDepartment of Industrial Engineering, Kasetsart University, Bangkok 10900, ThailandFlood forecasting is among the most important precaution measures to prevent devastating disasters affecting human life, properties, and the overall environment. It is closely involved with precipitation and streamflow data forecasting tasks. In this work, we introduced a multi-step discharge prediction framework based on deep learning models. A simple feature representation technique using a correlation of backward lags was enhanced with a time of concentration (T<sub>C</sub>) concept. Recurrent neural networks and their variants, coupled with the T<sub>C</sub>-related features, provided superior performance with over 0.9 Nash–Sutcliffe model efficiency coefficient and substantially high correlation values for multiple forecasted points. These results were consistent among both the Upper Nan and the Loei river basins in Thailand, which were used as case studies in this work.https://www.mdpi.com/2073-4441/14/18/2898discharge predictionflood forecastingtime of concentrationdeep learningrecurrent neural networks |
| spellingShingle | Wandee Thaisiam Warintra Saelo Papis Wongchaisuwat Enhancing a Multi-Step Discharge Prediction with Deep Learning and a Response Time Parameter Water discharge prediction flood forecasting time of concentration deep learning recurrent neural networks |
| title | Enhancing a Multi-Step Discharge Prediction with Deep Learning and a Response Time Parameter |
| title_full | Enhancing a Multi-Step Discharge Prediction with Deep Learning and a Response Time Parameter |
| title_fullStr | Enhancing a Multi-Step Discharge Prediction with Deep Learning and a Response Time Parameter |
| title_full_unstemmed | Enhancing a Multi-Step Discharge Prediction with Deep Learning and a Response Time Parameter |
| title_short | Enhancing a Multi-Step Discharge Prediction with Deep Learning and a Response Time Parameter |
| title_sort | enhancing a multi step discharge prediction with deep learning and a response time parameter |
| topic | discharge prediction flood forecasting time of concentration deep learning recurrent neural networks |
| url | https://www.mdpi.com/2073-4441/14/18/2898 |
| work_keys_str_mv | AT wandeethaisiam enhancingamultistepdischargepredictionwithdeeplearningandaresponsetimeparameter AT warintrasaelo enhancingamultistepdischargepredictionwithdeeplearningandaresponsetimeparameter AT papiswongchaisuwat enhancingamultistepdischargepredictionwithdeeplearningandaresponsetimeparameter |