Modeling of a real industrial wastewater treatment plant based on aerated lagoon using a neuro-evolutive technique

Aerated lagoons are biological systems used for the treatment of different types of wastewaters and many operating parameters influence the performance of these systems. Thus, in the current study, an industrial aerated lagoon system was modeled in terms of the operating parameters with the goal of predicting its performances under different operating conditions in order to highlight possible bottlenecks or potential improvements. For this purpose, a neuro-evolutive approach, combining differential evolution (DE) algorithm and artificial neural networks (ANN), was employed. Two DE variants based on opposition-based learning and of chaos theory were used to determine the optimal models and to perform a process optimization. Multiple models in various configurations (simple or organized in stacks, with single or multiple outputs) were determined. The mean squared error of the best solutions were in the order of 10(-4), illustrating a good agreement between the model predictions and experimental data and demonstrating the effectiveness and reliability of the developed models. (C) 2020 Institution of Chemical Engineers. Published by Elsevier B.V. All rights reserved.

Automated summary

In this study, an industrial aerated lagoon system was modeled using a neuro-evolutive approach to predict its performance and identify potential improvements under different operating conditions. The models, based on a combination of differential evolution algorithm and artificial neural networks, showed good agreement with experimental data, demonstrating their effectiveness and reliability in optimizing the treatment process.