Open Access Open Access  Restricted Access Subscription Access
Open Access Open Access Open Access  Restricted Access Restricted Access Subscription Access

Applications of Artificial Neural Network for Streamflow Forecasting-A Review


Affiliations
1 Department of Civil Engineering, G. Pulla Reddy Engineering College Kurnool, India
2 Department of Civil Engineering S.V.U. College of Engineering Tirupati – 517 502 Andhra Pradesh, India
3 Department of Civil Engineering, Sree Vidyanikethan Engineering College, A. Rangampet Tirupati – 517 102 Andhra Pradesh, India
     

   Subscribe/Renew Journal


Estimating streamflow is important in determining the water resource availability and assessing the flood, drought management and mitigation studies. Continuous investigation of streamflow history and monitoring of streamflow data is an effective way to establish a reliable forecast. These forecasting requires long length of data to analysis. The analysis can be done based on traditional methods. These methods require, more number of data, time consuming and tedious process. Therefore, these forecasts can hamper the development and management of water managers or authorities to effective utilization of water resources in a suitable manner. Therefore, there is a need of the hour to search alternative methods for the reliable forecasts. Data driven models such as Artificial Neural Networks (ANN) have proven to be an efficient alternative to traditional methods for assessing and modeling quantitative and qualitative in the domain of water resources engineering and management. Therefore, in the present paper an attempt have been made to investigate to study the applications of ANN in streamflow forecasting. Selected ANNs applications are only reviewed in the current paper. Soft computing tools are becoming popular in solving hydrological problems. Among the various soft computing methods ANN tools have immense strength to deal with such complex problems and becoming promising tools due to their ability in modelling of nonlinear process. This study will be helpful to enhance the frontiers for new research in the domain of hydrology. Further future research need to be explored towards the extraction of the knowledge that is contained the connection weights of the selected trained ANN models and also researchers should focus on selection of optimal number of input for the development of ANN models.


Keywords

Artificial Neural Networks, Streamflow, Time Series.
User
Subscription Login to verify subscription
Notifications
Font Size


  • Applications of Artificial Neural Network for Streamflow Forecasting-A Review

Abstract Views: 335  |  PDF Views: 2

Authors

M. Basha Mohiddin
Department of Civil Engineering, G. Pulla Reddy Engineering College Kurnool, India
P. Mallikarjuna
Department of Civil Engineering S.V.U. College of Engineering Tirupati – 517 502 Andhra Pradesh, India
Sreenivasulu Dandagala
Department of Civil Engineering, Sree Vidyanikethan Engineering College, A. Rangampet Tirupati – 517 102 Andhra Pradesh, India

Abstract


Estimating streamflow is important in determining the water resource availability and assessing the flood, drought management and mitigation studies. Continuous investigation of streamflow history and monitoring of streamflow data is an effective way to establish a reliable forecast. These forecasting requires long length of data to analysis. The analysis can be done based on traditional methods. These methods require, more number of data, time consuming and tedious process. Therefore, these forecasts can hamper the development and management of water managers or authorities to effective utilization of water resources in a suitable manner. Therefore, there is a need of the hour to search alternative methods for the reliable forecasts. Data driven models such as Artificial Neural Networks (ANN) have proven to be an efficient alternative to traditional methods for assessing and modeling quantitative and qualitative in the domain of water resources engineering and management. Therefore, in the present paper an attempt have been made to investigate to study the applications of ANN in streamflow forecasting. Selected ANNs applications are only reviewed in the current paper. Soft computing tools are becoming popular in solving hydrological problems. Among the various soft computing methods ANN tools have immense strength to deal with such complex problems and becoming promising tools due to their ability in modelling of nonlinear process. This study will be helpful to enhance the frontiers for new research in the domain of hydrology. Further future research need to be explored towards the extraction of the knowledge that is contained the connection weights of the selected trained ANN models and also researchers should focus on selection of optimal number of input for the development of ANN models.


Keywords


Artificial Neural Networks, Streamflow, Time Series.

References