Deep Learning with a Long Short-Term Memory Networks Approach for Rainfall-Runoff Simulation
Considering the high random and non-static property of the rainfall-runoff process, lots of models are being developed in order to learn about such a complex phenomenon. Recently, Machine learning techniques such as the Artificial Neural Network (ANN) and other networks have been extensively used by...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2018-10-01
|
| Series: | Water |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2073-4441/10/11/1543 |
| _version_ | 1811261452110004224 |
|---|---|
| author | Caihong Hu Qiang Wu Hui Li Shengqi Jian Nan Li Zhengzheng Lou |
| author_facet | Caihong Hu Qiang Wu Hui Li Shengqi Jian Nan Li Zhengzheng Lou |
| author_sort | Caihong Hu |
| collection | DOAJ |
| description | Considering the high random and non-static property of the rainfall-runoff process, lots of models are being developed in order to learn about such a complex phenomenon. Recently, Machine learning techniques such as the Artificial Neural Network (ANN) and other networks have been extensively used by hydrologists for rainfall-runoff modelling as well as for other fields of hydrology. However, deep learning methods such as the state-of-the-art for LSTM networks are little studied in hydrological sequence time-series predictions. We deployed ANN and LSTM network models for simulating the rainfall-runoff process based on flood events from 1971 to 2013 in Fen River basin monitored through 14 rainfall stations and one hydrologic station in the catchment. The experimental data were from 98 rainfall-runoff events in this period. In between 86 rainfall-runoff events were used as training set, and the rest were used as test set. The results show that the two networks are all suitable for rainfall-runoff models and better than conceptual and physical based models. LSTM models outperform the ANN models with the values of <inline-formula> <math display="inline"> <semantics> <msup> <mi>R</mi> <mn>2</mn> </msup> </semantics> </math> </inline-formula> and <inline-formula> <math display="inline"> <semantics> <mrow> <mi>N</mi> <mi>S</mi> <mi>E</mi> </mrow> </semantics> </math> </inline-formula> beyond 0.9, respectively. Considering different lead time modelling the LSTM model is also more stable than ANN model holding better simulation performance. The special units of forget gate makes LSTM model better simulation and more intelligent than ANN model. In this study, we want to propose new data-driven methods for flood forecasting. |
| first_indexed | 2024-04-12T19:04:39Z |
| format | Article |
| id | doaj.art-b05f11db694e4d7ebc3ddb033480ec06 |
| institution | Directory Open Access Journal |
| issn | 2073-4441 |
| language | English |
| last_indexed | 2024-04-12T19:04:39Z |
| publishDate | 2018-10-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Water |
| spelling | doaj.art-b05f11db694e4d7ebc3ddb033480ec062022-12-22T03:20:04ZengMDPI AGWater2073-44412018-10-011011154310.3390/w10111543w10111543Deep Learning with a Long Short-Term Memory Networks Approach for Rainfall-Runoff SimulationCaihong Hu0Qiang Wu1Hui Li2Shengqi Jian3Nan Li4Zhengzheng Lou5School of Water Conservancy and Environment, Zhengzhou University, Zhengzhou 450001, ChinaSchool of Water Conservancy and Environment, Zhengzhou University, Zhengzhou 450001, ChinaSchool of Information Engineering, Zhengzhou University, Zhengzhou 450001, ChinaSchool of Water Conservancy and Environment, Zhengzhou University, Zhengzhou 450001, ChinaSchool of Water Conservancy and Environment, Zhengzhou University, Zhengzhou 450001, ChinaSchool of Information Engineering, Zhengzhou University, Zhengzhou 450001, ChinaConsidering the high random and non-static property of the rainfall-runoff process, lots of models are being developed in order to learn about such a complex phenomenon. Recently, Machine learning techniques such as the Artificial Neural Network (ANN) and other networks have been extensively used by hydrologists for rainfall-runoff modelling as well as for other fields of hydrology. However, deep learning methods such as the state-of-the-art for LSTM networks are little studied in hydrological sequence time-series predictions. We deployed ANN and LSTM network models for simulating the rainfall-runoff process based on flood events from 1971 to 2013 in Fen River basin monitored through 14 rainfall stations and one hydrologic station in the catchment. The experimental data were from 98 rainfall-runoff events in this period. In between 86 rainfall-runoff events were used as training set, and the rest were used as test set. The results show that the two networks are all suitable for rainfall-runoff models and better than conceptual and physical based models. LSTM models outperform the ANN models with the values of <inline-formula> <math display="inline"> <semantics> <msup> <mi>R</mi> <mn>2</mn> </msup> </semantics> </math> </inline-formula> and <inline-formula> <math display="inline"> <semantics> <mrow> <mi>N</mi> <mi>S</mi> <mi>E</mi> </mrow> </semantics> </math> </inline-formula> beyond 0.9, respectively. Considering different lead time modelling the LSTM model is also more stable than ANN model holding better simulation performance. The special units of forget gate makes LSTM model better simulation and more intelligent than ANN model. In this study, we want to propose new data-driven methods for flood forecasting.https://www.mdpi.com/2073-4441/10/11/1543LSTMrainfall-runoffflood events |
| spellingShingle | Caihong Hu Qiang Wu Hui Li Shengqi Jian Nan Li Zhengzheng Lou Deep Learning with a Long Short-Term Memory Networks Approach for Rainfall-Runoff Simulation Water LSTM rainfall-runoff flood events |
| title | Deep Learning with a Long Short-Term Memory Networks Approach for Rainfall-Runoff Simulation |
| title_full | Deep Learning with a Long Short-Term Memory Networks Approach for Rainfall-Runoff Simulation |
| title_fullStr | Deep Learning with a Long Short-Term Memory Networks Approach for Rainfall-Runoff Simulation |
| title_full_unstemmed | Deep Learning with a Long Short-Term Memory Networks Approach for Rainfall-Runoff Simulation |
| title_short | Deep Learning with a Long Short-Term Memory Networks Approach for Rainfall-Runoff Simulation |
| title_sort | deep learning with a long short term memory networks approach for rainfall runoff simulation |
| topic | LSTM rainfall-runoff flood events |
| url | https://www.mdpi.com/2073-4441/10/11/1543 |
| work_keys_str_mv | AT caihonghu deeplearningwithalongshorttermmemorynetworksapproachforrainfallrunoffsimulation AT qiangwu deeplearningwithalongshorttermmemorynetworksapproachforrainfallrunoffsimulation AT huili deeplearningwithalongshorttermmemorynetworksapproachforrainfallrunoffsimulation AT shengqijian deeplearningwithalongshorttermmemorynetworksapproachforrainfallrunoffsimulation AT nanli deeplearningwithalongshorttermmemorynetworksapproachforrainfallrunoffsimulation AT zhengzhenglou deeplearningwithalongshorttermmemorynetworksapproachforrainfallrunoffsimulation |