Approximating of BOI parameter from KOI parameter using artificial neural networks

Keywords: Biological Waste water Treatment Plant, parameter, Artificial Neural Networks (ANN), Chemical Oxygen Demand (COD), Biological Oxygen Demand (BOD5). The Artificial Neural Networks is a sub-branch of the Artificial Intelligence Science and artificial systems which take the assumed working principle of the human brain as a principle. ANN is one of the most popular subjects of the modern science with learning capability, adaptation, characteristics of working with the minimum information, rapid working and recognition convenience. The purpose in this thesis study is estimating the Biological Oxygen Demand (BOD5) with the Artificial Neural Networks (ANN) by means of using the experimental Chemical Oxygen Demand (COD) which is measures in a Waste water Treatment Plants A and B. To this end, first of all, ANN was trained and tested by means of using 365 experimental data which are gathered together from a waste water treatment plant A and ANN was trained and tested by means of using 365 experimental data which are gathered together from a waste water treatment plant B. ANN is a data processing system which is distributed as parallel and composed of a lot of process elements and connections. The unique feature of ANN lies in learning the relationship between the inputs and outputs of experimental data without need of any suggestion and assumption and making generalization accordingly. The theoretical results and experimental results obtained from ANN are compared. According to comparison results, ANN algorithm which is developed based on MATLAB could be an alternative method which could be used in estimating the performance parameters in the Waste water Treatment Plants.