:: Volume 2, Issue 2 (Spring 2015) ::
3 2015, 2(2): 1-12 Back to browse issues page
Estimation of nitrate concentrations in well and spring water using ANFIS and SVM models
Mojtaba Khoshravesh , Mohammad-Ali Gholami , Jamal Abbas-palangi , Mohammad Mirnaseri
Abstract:   (1369 Views)

Groundwater contamination by nitrate is a globally growing problem due to the population growth and increase of demand for food supplies.Increasing nitrate concentrations in soil solution and leaching into the ground water table cause water pollution and disturbs the ecological balance. In addition to natural nitrogen cycle, nitrate can be entered to soil and water from the human waste, urban and industrial wastes andagricultural activities and cause undesirable effects on human health.Because of Iran's semi-arid climate, groundwater is one of the most important water resources.Ground water supplies are always at risk of nitrate contamination.An adaptive network-based fuzzy inference system (ANFIS)and support vector machine (SVM) as models that analyze the available information, provide the possibility of extracting nonlinear and unknown relationships, and haveapplied successfully in many branches of science, especially in water science.In this study, intelligent techniques such as ANFIS and SVM used as a quite flexible tool in simulation of nitrate concentration changes in well and spring waters.The advantage of this approach is high flexibility of intelligent systems against complex functions and use of inputs that are readily available. The results showed that the simulation of nitrate using the ANFIS system was better than the SVM and also this simulation for spring water was better than the simulation for well water.Determination coefficient of ANFIS method was obtained as 0.93 and 0.95 for well and spring waters, respectively. Furthermore, determination coefficient of SVM method was obtained as 0.88 and 0.91 for well and spring waters, respectively.

Keywords: Nitrate pollution, Support vector machine, ANFIS model, Well water, Spring water.
Full-Text [PDF 444 kb]   (535 Downloads)    
Type of Study: Research | Subject: Groundwater
Received: 2016/02/8 | Accepted: 2016/09/29 | Published: 2016/12/1

XML     Print

Volume 2, Issue 2 (Spring 2015) Back to browse issues page