ANN-based predictive model for performance evaluation of paper and pulp effluent treatment plant

Abstract

Neural networks (NNs) have been widely used for complex processes that are poorly described by first principle models, such as wastewater biological treatment systems. In this paper, we propose an Artificial Neural Network (ANN) based predictive model for assessing the performance of paper and pulp effluent treatment plants. Mathematical models were created for the thickener area of the clarifier by correlating process control parameters such as mean cell residence time (θc), initial suspended solid concentration (Co), underflow concentration (Cu) and recycling ratio (R). For any values of Cu, Co, R and θc, area of the secondary clarifier can be determined using the model developed based on ANN. The predicted models give a rational approach to the design of secondary clarifier. The developed models prove consistently well in the face of varying accuracy and size of input data phase.