IJISAE, 2015, 3(4), 131-135 | 131 An Artificial Neural Network Model for Wastewater Treatment Plant of Konya Abdullah Erdal TÜMER* 1 , Serpil EDEBALİ 2 Accepted 15 th August 2015 DOI: 10.18201/ijisae.65358 DOI: 10.1039/b000000x Abstract: In this study, modelling of Konya wastewater treatment plant was studied by using artificial neural network with different architectures in Matlab software. All data were obtained from wastewater treatment plant of Konya during daily records over four month. Treatment efficiency of the plant was determined by taking into account of input values of pH, temperature, COD, TSS and BOD with output values TSS. Performance of the model was compared via the parameters of Mean Squared Error (MSE), and correlation coefficient (R). The suitable architecture of the neural network model is determined after several trial and error steps. According to the modelling study, the ANN can predict the plant performance with correlation coefficient (R) between the observed and predicted output variable reached up to 0.96. Keywords: Artificial Neural Network, Modelling, Wastewater Treatment Plant, Performance. 1. Introduction In last two decades, Artificial Neural Network (ANN) methods have been applied to various areas of environmental issues such as wastewater treatment. The wastewater treatment process is quite complex. However the developments in intelligent methods make them possible to use in complex systems modelling [1]. It can be used for better prediction of the process performance owing to their high accuracy, adequacy and quite promising applications in engineering [2,3,4]. There are certain key descriptions of variables which can be used to assess the wastewater treatment plant performance. These variables consist of Chemical Oxygen Demand (COD), Biological Oxygen Demand (BOD) and Total Suspended Solids (TSSs). Until now, most of the available studies for modelling Waste Water Treatment Plants (WWTPs) used these variables. The ANN based models find out very satisfactory results. Some of these models based on ANNs are as follows. ANN model was developed for BOD removal process in horizontal subsurface flow constructed wetlands by Akratos et al. [5]. Farouq S et al. in [6] used neural network with single input and multi-input layers and gave comparable predictions of the plant performance criteria. To prediction of biological oxygen demand and suspended solid concentrations based on ANN were presented by Hamed et al. [7]. Total suspended solid (TSS) is an indication of plant performance. A simple prediction models based on neural network for TSS was demonstrated in [8]. Many other ANNs model for wastewater treatment have been proposed either in the past [9-13] or more recently [14- 18]. In this work, an effort has been made to model the wastewater treatment data using the artificial neural networks to predict the performance of Konya Wastewater Treatment Plant. The data set was consist of flow rate, COD, BOD, TSS and ph at 25 0C temperature as input layer variables and TSS as output layer variables. The samples collected on a day during for a period of 4 month. The results of the modelling study expressed briefly high correlation coefficient (R) between the measured and predicted output variables reaching up to 0.9. 2. Materials and Methods 2.1. Artificial Neural Network Artificial Neural Network (ANN) is an information processing system derived from biological nervous systems of brain. The goal of an ANN is to compute between output and input values with some internal calculations [19]. There is a two-stage operation mode of artificial neural networks. One of them is training the other testing stage. Once it must be trained to use an artificial neural network. The training is carried out using some of the inputs and outputs data set. ANN makes a generalization of these data. Artificial neural networks consists of three layers, including inputs, output and hidden layers, and there are many neurons in each layer as shown (Figure.1). _______________________________________________________________________________________________________________________________________________________________ 1 Computer Engineering Department, University of Necmettin Erbakan, Konya, Turkey 2 Chemical Engineering Department, University of Selçuk, Konya, Turkey * Corresponding Author: Email: tumer@konya.edu.tr Note: This paper has been presented at the International Conference on Advanced Technology&Sciences (ICAT'15) held in Antalya (Turkey), August 4-7, 2015 International Journal of Intelligent Systems and Applications in Engineering ISSN:2147-67992147-6799 www.ijisae.org Original Research Paper International Journal of Intelligent Systems and Applications in Engineering