Application of Soft Compiting Techniques in River Flow Modeling in The
Case of Euphrates-Tigres Basin
Abstract
River stream estimation is a subject matter that needs constant research
and development since it is all-important in the management of water
resources, meeting the water demand, irrigation and agricultural
activities, and providing distant signal in unwanted situations such as
floods. Unfortunately, a universal technique has not been found yet
although many techniques have been used for estimation and modelling.
This has made it necessary to develop different techniques and/ or to
make comparisons between techniques and to determine the most accurate
method for the parameters used. In this study, using the 1981-2010 flow
data of 14 stations located across Euphrates-Tigris basin, evaluations
have been made through Adaptive-Network Based Fuzzy Inference Systems
(ANFIS), Support Vector Regression (SVR-SVMR) techniques, and the newly
used Gauss Process Regression (GPR), Extreme Learning Machine (ELM) and
Emotional Neural Network (ENN) artificial intelligence techniques, and
through rank analysis, it is aimed to find out which technique gives
better results and to overcome some problems in traditional methods.
Although all models work well, the sequence with regards to the
comparison outcomes of the techniques obtained from rank analysis was
observed to be ELM, GPR, ENN, SVM, ANFIS respectively. In addition,
stream values were used in the whole study, these values were examined
within 3 different combinations and it was observed that the best result
was found in the combination of
[input]Q(t-3),Q(t-2),Q(t-1)/[output]Q(t). Keywords: River Flow
Modelling; ANFIS; SVM; GPR; ELM; ENN