A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
EXTENSION OF STREAM FLOW SERIES USING ARTIFICIAL NEURAL NETWORKS
Sufficient length of record of flow is very important for accurate estimation of probable maximum flood of a river at the location of any hydraulic structure for its economic design or planning of reservoir operation policy. Many a times the length of record of flow is insufficient or not available at all. Normally under such situations hydrologists either extend the short-term data somehow or generate the data with the properties of historical observed data. The most accepted method to extend short- term data with the help of long-term data at a neighboring station is Langbein's log deviation. This method consists of developing a correlation between the streamflows at short-term and long-term gauging stations and to estimate missing flow series the procedure requires cumbersome mathematical computations. This paper demonstrates the application of artificial neural networks (ANNs) in estimating the missing mean monthly streamflow of Narmada river at Dindori G & D site (M.P.), India using the monthly flow data obtained from the Cental Water Commission. The performance of the artificial neural network is compared with that of the Langbein's log deviation method. The results suggest that the ANN approach may provide a superior alternative to Langbein's log deviation method for estimating the missing flow series or forecasting flows with less computational effort and time.
EXTENSION OF STREAM FLOW SERIES USING ARTIFICIAL NEURAL NETWORKS
Sufficient length of record of flow is very important for accurate estimation of probable maximum flood of a river at the location of any hydraulic structure for its economic design or planning of reservoir operation policy. Many a times the length of record of flow is insufficient or not available at all. Normally under such situations hydrologists either extend the short-term data somehow or generate the data with the properties of historical observed data. The most accepted method to extend short- term data with the help of long-term data at a neighboring station is Langbein's log deviation. This method consists of developing a correlation between the streamflows at short-term and long-term gauging stations and to estimate missing flow series the procedure requires cumbersome mathematical computations. This paper demonstrates the application of artificial neural networks (ANNs) in estimating the missing mean monthly streamflow of Narmada river at Dindori G & D site (M.P.), India using the monthly flow data obtained from the Cental Water Commission. The performance of the artificial neural network is compared with that of the Langbein's log deviation method. The results suggest that the ANN approach may provide a superior alternative to Langbein's log deviation method for estimating the missing flow series or forecasting flows with less computational effort and time.
EXTENSION OF STREAM FLOW SERIES USING ARTIFICIAL NEURAL NETWORKS
Singh, K. K. (author) / Kumar, Suresh (author)
ISH Journal of Hydraulic Engineering ; 13 ; 55-65
2007-01-01
11 pages
Article (Journal)
Electronic Resource
Unknown
Stream Flow Modeling of River Swat Using Regression and Artificial Neural Networks (ANNs) Techniques
| British Library Conference Proceedings | 2011
River flow forecasting using artificial neural networks
| British Library Conference Proceedings | 2004
PREDICTION OF DISPERSION COEFFICIENTS IN A SMALL STREAM USING ARTIFICIAL NEURAL NETWORKS
| British Library Conference Proceedings | 2007
River Flow Modeling Using Artificial Neural Networks
| Online Contents | 2004
River Flow Modeling Using Artificial Neural Networks
| British Library Online Contents | 2004