Application of a neural network for modeling and identification in a wastewater treatment plant

Abstract This work presents the modeling of a wastewater treatment plant using nonlinear ordinary differential equations that allow dissolved oxygen to be monitored. The plant configuration is based on an anoxic bioreactor, an aerobic bioreactor and a secondary settler connected in cascade. The mathematical modeling of the plant based on the ASM model No. 1. On the other hand, it is possible to model the system by neural model of the plant using hierarchical neural networks. The objective of which is to identify some reactions that occur in bioreactors. The implementation of the identification of reactions is carried out through numerical simulations that allow validating the effectiveness of the proposed model and also obtaining relevant results to be able to apply them in future work in neural control.