|
PRINT ISSN : 2319-7692
Online ISSN : 2319-7706 Issues : 12 per year Publisher : Excellent Publishers Email : editorijcmas@gmail.com / submit@ijcmas.com Editor-in-chief: Dr.M.Prakash Index Copernicus ICV 2018: 95.39 NAAS RATING 2020: 5.38 |
The accelerated erosion and the sediment outflow from agricultural lands is a serious global problem. Mankind will be facing great challenges in the next few decades. The present study was undertaken to estimate the suspended sediment concentration using Fuzzy Logic (FL), Multiple Linear Regression (MLR) and Sediment Rating Curve (SRC) models for the Vamsadhara river catchment comprising of 7820 km2 Area, situated between Mahanadi and Godavari river basins in south India. The perfect input of FL, MLR and SRC models was found by the use of Gamma Test (GT). Three different types of performance indicators viz. root mean square error (RMSE), correlation coefficient (r) and coefficient of efficiency (CE) were used to evaluate the accuracy of various models. Based on the performance analysis SRC, MLR and FL models were used for comparison. Comparison of training period length was also made utilizing two different architectures. Daily simulations using inputs with architecture (3*1) three years of training and one year of testing (RMSE-104.85 kg/sec, r-0.963 and CE value 87.93%) performed better than the simulation with architecture (2*2) two years of training and two years of testing. The study demonstrates that fuzzy logic model utilizing Gamma Test for input choosing has superior performance for sediment yield estimation in comparison to the conventional models such as MLR and SRC. The results indicated that fuzzy logic can be applied successfully to provide high accuracy and reliability for sediment estimation.