Research Information System
Innovation, Sustainability, Technology and Education in Civil Engineering Conference, Hatay, Turkey, 13 - 15 June 2019, vol.-, pp.630-639
In this study, it was aimed to estimate the manganese(II) ion (Mn2+) concentration from in situ turbidity measurements from the dam. The Mn2+ concentration and turbidity were monitored daily at the drinking water treatment plants of the Sinop Municipality during a period of one year from 2014 to 2015. The conventional regression analysis (CRA) and artificial neural networks (ANNs) were used for estimation of Mn2+ concentration. The CRA was performed with four different equations. Linear, quadratic, exponential, and power functions were applied to the training of the models. In addition to different learning rate and momentum, two different modeling algorithms, namely Levenberg-Marquardt and gradient descent with momentum were used in the ANNs. In all models data were divided into training, validation, and test data sets. The estimation results were compared with each other employing root mean squared error, mean absolute error, and determination coefficients. It was concluded that the CRA and ANNs can be used as a convenient method for estimating Mn2+ concentration in a fast and cost-effective manner with turbidity.