Author(s)

V. Tomenko, S. Ahmed & V. Popov

Abstract

Artificial neural network (ANN) models were developed to predict the performance of a constructed wetland wastewater treatment plant (CWWTP). The model assesses the Biochemical Oxygen Demand (BOD) concentration at outlet of a treatment plant. Training of ANN models was based on experimental results of a pilot plant study in India. The data used in this work were obtained under various hydraulic and BOD loading. Regular records of BOD were made at inlet, and outlet levels through various stages of the treatment process for over 18 months. The ANN-based models were found to provide an efficient and a robust tool in predicting CWWTP performance. Keywords: neural networks, constructed wetland, model studies, prediction, optimization, biochemical oxygen demand. 1 Introduction The proper operation and management of constructed wetland wastewater treatment plants (CWWTP) is receiving attention because of the rising concern about environmental issues and growing importance of sustainable and natural wastewater treatment techniques. Improper design and operation of a CWWTP may cause serious environmental and public health implications, as its effluent may contaminate receiving water body, causing severe aquatic pollution and spread various water born diseases. For proper design and assessment of quality of non-conventional wastewater treatment and thereafter to conserve the receiving water bodies, reliable prediction of effluent from Constructed Wetland (CW) is essential. A better control can be achieved by developing a

Keywords

neural networks, constructed wetland, model studies, prediction, optimization, biochemical oxygen demand.