Application of an Artificial Intelligence to the Estimation of Water Quality Parameters: Water Quality of Nigde Creek Water, Turkey

Abstract

Artificial intelligence is used for the estimation of water quality parameters of Nigde creek in the central part of Turkey. Nigde creek basin contains the two largest region of Turkey which is Nigde city and Bor district. Moreover, it is the major agricultural area in the region as well as highly agricultural and industrialized. In this study, analyses have been conducted on water samples taken from five stations located along the Nigde creek. Hydro-chemical analyses have been conducted on the water samples, the results have been evaluated according to water quality parameters. The laboratory data with an artificial neural network computer program called clean water quality program-CWQP, which is designed according to water quality parameters, its place has been defined among the classifications of water quality. Quality parameters used have been prepared in compliance to the In-continent surface water classification as stated in the regulation of water pollution and checking for Turkey. Accordingly, water quality has been classified in four groups as Quality I, II, III and IV. Selection of appropriate input parameters to estimate another one is key to use ANN. This is essential to obtain maximum success with minimum error. The input parameters selected must be those which affect most the output one. In this study, therefore, water quality at five stations has been determined to be of Class IV water quality with CWQP. Furthermore, lead has stepped up as the most important polluter in the analyses, which has the negative effects on water quality in all stations and precautions are needed to be taken to protect Nigde creek. Results obtained from the CWQP have been discovered to be indifferent from the ones obtained through the classical methods, implying their availability for use comfortably in water pollution areas. This program has ensured rapid results and non-specialized study in determining the water quality.