stimation of Some Ions Base on Electrical Conductivity Using Adaptive Neuro-Fuzzy Inference System in Grounwater

Document Type : Research

Authors

1 Islamic Azad University, Hamedan Branch, Club of Young and Elite Researchers, Hamedan, Iran

2 Ph.D. student of water resources, Department of Water Engineering, Faculty of Agriculture, Bu Ali Sina University, Hamedan

10.22092/wmej.2016.113501

Abstract

In evaluating water quality for measure each of the ion, chemicals material and time is consuming, while measuring the electrical conductivity by conductivity meter is easily. Therefore In order to estimate of the Ca2+, Mg2+, Na+, SO4- and Cl- base on electrical conductivity and developing a simple and rational relationship between some of cations, anions and electrical conductivity at hamedan-bahar plain and comparisons to adaptive neuro-fuzzy inference system (ANFIS) a study was conducted in year of 2002. The results of this study comparisons with regression models. Determine the optimum Functional relationships parameters by correlation coefficient between data and RMSE was used. The comparisons of regressions and ANFIS results indicate that ANFIS performs more accurate of cation and anion estimates. Base of ANFIN model, the values of R2 for Ca, Mg, Na, Cl and SO4 were 0.60, 0.71, 0.86, 0.84 and 0.73. Also, the results showed that the best and worst of estimation between cations, anions and electrical conductivity for Na and Ca were observed, respectively. The analysis of output results from ANFIS model shown that this model has a high accuracy and ability to estimation of cation and anion in hamedan-bahar plain.

Keywords

Files

History

  • Receive Date: 21 February 2014
  • Revise Date: 15 May 2014
  • Accept Date: 21 June 2014

Share

How to cite

Statistics