Document Type : Original Research Paper

Authors

• A.A Nadiri ¹
• F sadeghi Aghdam ²
• A Asgharai Moghaddam ³

¹ Assistant Professor, Department of Geosciences, Faculty of Natural science, Tabriz University, Tabriz, Iran

² Ph. D. Student, Department of Geosciences, Faculty of Natural science, Tabriz University, Tabriz, Iran

³ Professor, Department of Geosciences, Faculty of Natural science, Tabriz University, Tabriz, Iran

Abstract

This study presents an intelligence committee fuzzy logic (ICFL) model to estimate the concentration of total arsenic (III, V) in the area of the Sahand Dam basin, Iran. Because of a high concentration of arsenic (III, V) compared to WHO standard, Geolology Department of Tabriz University and East Azerbaijan Regional Water Authority have attempted to sampling and analysis of surface water and groundwater resources in the mentioned area. Hydrochemical parameters Including pH, SO₄^2-, NO^3--N, F^-, Fe(II, III) and As used as input parameters for Mamdani fuzzy logic (MFL), Larsen fuzzy logic (LSL) and Sugeno fuzzy logic (SFL) to estimate arsenic concentrations. The results in train and test steps showed that all of these models have a similar fitting to the arsenic concentration data in the study area. The ICFL model was adopted to combine the output of the three single fuzzy models instead of the selecting superior single model. To reap advantage of all three models, the weighted combination of the output of fuzzy models used to create a committee fuzzy model. The mentioned model uses particles population algorithm, to obtain weight based on the output of the three fuzzy models to estimate the total arsenic concentration. The results of ICFL model shows, significant fitting improvement compare to individual fuzzy logic models.

Keywords

• ARSENIC
• Sahand Dam
• Sugeno fuzzy logic
• Larsen fuzzy logic
• Mamdani fuzzy logic
• Intelligence committee fuzzy logic