Please use this identifier to cite or link to this item: http://localhost:8081/xmlui/handle/123456789/7169
Title: ESTIMATION OF SEDIMENT YIELD USING ARTIFICIAL NEURAL NETWORK
Authors: Singh, Abhishek Kumar
Keywords: CIVIL ENGINEERING;SEDIMENT YIELD;ARTIFICIAL NEURAL NETWORK;SPUR
Issue Date: 2001
Abstract: Evaluation of sediment yield from a watershed is important for many reasons. Deposition of excess sediment in reservoir reduces it's capacity, thereby adversely affecting the water supply for irrigation, domestic, industrial use and power generation .The deposition of sediments on river beds and banks causes braiding of river reaches and submergence of flood plains of rivers during floods. Many sediment yield predictive models like USLE, MUSLE, CREAMS, SPUR and regression equations are known for estimating sediment yield. But most of these models are site and situation specific; hence have limited use. Considering the results of Artificial Neural Networks (ANN) in many related fields, in the present work an attempt has been made to use ANN for predicting sediment yield from a watershed. ANN tries to approximate a transfer function that transforms a bounded input vector into a bounded output vector. In the present case, feed forward networks with error back propagation learning rule have been used for training with the data available for the 100 watersheds. For this five input parameters namely drainage area, land slope, land cover factor, drainage density, average annual rainfall have been used. Keeping in view the overall efficiency, A Network consisting of one hidden layer having four nodes have been finalized. This network was trained. Using the developed and trained network, sediment yield from a small watershed has been computed and compared with the results from an existing model. The result from ANN was found to be very satisfactory.
URI: http://hdl.handle.net/123456789/7169
Other Identifiers: M.Tech
Research Supervisor/ Guide: Dubey, Prakash
Chandra, A. M.
metadata.dc.type: M.Tech Dessertation
Appears in Collections:MASTERS' THESES (Civil Engg)

Files in This Item:
File Description SizeFormat 
CED G10404.pdf3.13 MBAdobe PDFView/Open


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.