Evaluation of Intelligent Model in Estimating Electrical Conductivity in Groundwater(Case study: Rayen plain )

In this study, effectiveness of the intelligent method of ANFIS-FCM adaptive fuzzy neural inference system to predict the electrical conductivity in groundwater due to physical and chemical parameters in the Rayen plain was evaluated. To achieve this, 29 water samples were taken from wells and springs across the plain and the data were chemically analyzed in the laboratory. Descriptive statistics of data and the correlation matrix of the studied parameters were obtained using SPSS software. By forming a correlation matrix, it was found that the potassium (K+), sodium (Na+), calcium (Ca2+), magnesium (Mg2+), chlorine (Cl-), sulfate (SO42-), total soluble solids (TDS), total hardness (TH), compared to other available parameters, have the highest correlation with electrical conductivity (EC). Therefore, the model inputs included the mentioned parameters and electrical conductivity was also selected as the output according to the research purpose. After standardization, the data were entered into MATLAB environment and the electrical conductivity of groundwater was predicted using ANFIS-FCM method. In this method, 70% of the data (20 samples) were selected as the training data set and 30% of the data (9 samples) for the test data set. For the training data set of ANFIS-FCM model, R2 and RMSE values were 0.99994, 0.0001569, respectively, and also for test data set of ANFIS-FCM model, 0.9844 and 0.041652 were resulted for R2 and RMSE, respectively. Using the results of this model, it was found that the estimated electrical conductivity in the study area had very good accuracy and high correlation with the measured values. As a result, the ANFIS-FCM intelligent method is an effective, efficient and accurate method to estimate the electrical conductivity of water.