A performance evaluation of neuro-fuzzy and regression methods in estimation of sediment load of river

Document Type : Research

Authors

1 Assistant Professor, Department of Watershed Management, Faculty of Agriculture and Natural Resources, Arak Branch, Islamic Azad University, Arak, Iran

2 Faculty member of Agricultural Education and Promotion Research Organization and Soil Protection and Watershed Management Research Institute

3 Faculty member of the Agricultural Research, Education and Promotion Organization

4 Faculty member of soil protection and watershed management research institute

5 Agricultural and Natural Resources Research Center of Markazei Province

10.22092/wmej.2016.109749

Abstract

Application of Neuro-fuzzy and tree regression models is not too old in hydrology of river sediments and also in watersheds and in this regard, sediment rating curves have been identified as the most common method for estimating sediment. In this study for comparison and correction of estimation methods of river sediment load, estimated rates of several uni-variate types of sediment rating curves and a multivariate type of sediment rating curves were investigated with Neuro-fuzzy and tree regression models in five selective hydro-metric station of different climatic zones of Iran and with various indexes of the accuracy and the precision.  The results show that the mean of the accuracy index of Neuro-fuzzy and tree regression models in selective stations are 151 and 536 percent respectively which shows low efficiency compared with sediment rating curves. The results of the application of multivariate sediment rating curve in various station shows that the rate of the accuracy index of multivariate sediment rating curve in the best case is belong to Glinak station with the rate of 1.12. Also the average value of the accuracy index of multivariate sediment rating curve is 1.15 which is an acceptable amount to the other considered various methods.

Keywords

Watershed Management Research
Volume 28, Issue 4
January 2016
Pages 104-113

Files

History

  • Receive Date: 04 April 2015
  • Revise Date: 20 September 2015
  • Accept Date: 22 October 2015

Share

How to cite

Statistics