A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
Forecasting monthly streamflows using heuristic models
Forecasting streamflow values is of great importance in hydrology and water resources engineering as it affects the water related inflow-demand management, dam structure design and river engineering studies. Apart from using some physics-based models of this parameter forecast, using the previously recorded streamflow values for forecasting the future values would be very interesting, as only streamflow time series will be needed there. The present study aimed at assessing three heuristic data driven approaches, namely, gene expression programming (GEP), support vector machine (SVM) and interactive trees (IT) in forecasting monthly streamflow records. Monthly data from Soofi-Chai river in Iran covering a period of 13 years were used and a local k-fold testing cross validation process was adopted for training and testing the applied models. The obtained results revealed that all the applied models could predict riverflow time series with good accuracy. The results also showed the importance of defining a through train-test block mode (here, k-fold testing) to get a better insight about the applied models.
Forecasting monthly streamflows using heuristic models
Forecasting streamflow values is of great importance in hydrology and water resources engineering as it affects the water related inflow-demand management, dam structure design and river engineering studies. Apart from using some physics-based models of this parameter forecast, using the previously recorded streamflow values for forecasting the future values would be very interesting, as only streamflow time series will be needed there. The present study aimed at assessing three heuristic data driven approaches, namely, gene expression programming (GEP), support vector machine (SVM) and interactive trees (IT) in forecasting monthly streamflow records. Monthly data from Soofi-Chai river in Iran covering a period of 13 years were used and a local k-fold testing cross validation process was adopted for training and testing the applied models. The obtained results revealed that all the applied models could predict riverflow time series with good accuracy. The results also showed the importance of defining a through train-test block mode (here, k-fold testing) to get a better insight about the applied models.
Forecasting monthly streamflows using heuristic models
Mohsenzadeh Karimi, Sahar (author) / Karimi, Sepideh (author) / Poorrajabali, Mohammad (author)
ISH Journal of Hydraulic Engineering ; 27 ; 73-78
2021-01-02
6 pages
Article (Journal)
Electronic Resource
Unknown
Streamflow , GEP , SVM , IT , cross validation
Forecasting of Monthly Streamflows Based on Artificial Neural Networks
| British Library Online Contents | 2009
Forecasting of Monthly Streamflows Based on Artificial Neural Networks
| Online Contents | 2009
Stochastic Generation of Monthly Streamflows
| ASCE | 2021
Modelling of Monthly Streamflows using Stochastic and ANN Models
| British Library Online Contents | 2011
SIMULATION OF MONTHLY STREAMFLOWS USING LINEAR REGRESSION AND ANN MODELS
| Taylor & Francis Verlag | 2009