|
Estimation of Suspended Sediment Load Using Meta-Heuristic Methods in Ahar Chai River
|
| |
|
|
|
Abstract: (8090 Views) |
| Correct estimation of suspended sediment and its impact on the design and management of water projects, have always vital roles in advancement of studies concerning the river engineering, especially with consideration of the technical and economic difficulties associated with the installation and operation of stations for measuring sedimentation. Therefore, presenting an appropriate strategy for accurate prediction of sediment load of rivers would be significantly valuable. The deficiency in having a full set of precise measurements for the influential parameters in the sedimentation process and also the complete non-linear nature of models for the corresponding methods result in the rather inaccurate estimation, and therefore without the possibility of evaluating the changes in the sedimentation carried by the flow as a function of time, it would be impossible to come up with a comprehensive model. The purpose of present study is the evaluation of capability of Artificial Neural Networks (ANN) and Genetic Programing (GP) in predicting the sediment load in the Ahar-Chai. In order to estimate the sediment load, flow rate data, precipitation data, temperature data, and earlier sedimentation data have been used in these models. These models were applied to the Ahar-Chai River located in the East-Azarbayjan Province and the results were investigated and were compared to the collected data. In order to assess the efficiency of each of aforementioned models, the calculated data using each model were compared to the observed data using parameters namely, Coefficient of determination (R2), Nash-Sutcliffe model efficiency coefficient (E) and root mean square error (RMSE). Finally, the Genetic Programing was identified as the best model in estimating the sedimentation in Ahar-Chai river was identified and recommended. |
|
| Keywords: Artificial Neural Network (ANN). Sediment load, Modeling, Ahar-Chai River, Genetic programing (GP) |
|
|
Full-Text [DOCX 327 kb]
(1560 Downloads)
|
Type of Study: Research |
Subject:
Special
Received: 2014/09/24 | Accepted: 2014/09/24 | Published: 2014/09/24
|
|
|
|
|
|
|
| Add your comments about this article |
|
|
|
