Neural network modeling of a complete ammonium nitrification system with parameter identification

Grechneva A.O., Tarasov V.D., Tarkhov D.A., Shatrov A.V.
 pdf (796K)

This paper explores the application of Physics-Informed Neural Networks for modeling the nitrification process in wastewater treatment. The relevance of this work stems from the need to improve treatment efficiency and enable real-time monitoring of parameter changes, which is challenging with traditional approaches such as Activated Sludge Models (ASM). The authors propose a method that combines solving the forward problem and identifying parameters of a system of differential equations describing nitrogen oxidation. The study conducts a comparative analysis of three neural network architectures: with a fixed identical number of neurons, with a fixed different number of neurons, and with a progressively growing number of neurons during training. Multicriteria optimization based on the Pareto front construction was employed to find optimal parameters (number of neurons and the loss function weight coefficient). The results demonstrate that the proposed approach effectively reconstructs the dynamics of nitrogen compound concentrations with high accuracy. The architecture with a fixed different number of neurons, combined with optimally selected weight coefficients, provided the best agreement with experimental data while maintaining an acceptable residual of the equations.

Keywords: physics-informed neural networks, PINN, wastewater treatment, nitrification, parameter identification, modeling, machine learning, multi-criteria optimization
Citation in English: Grechneva A.O., Tarasov V.D., Tarkhov D.A., Shatrov A.V. Neural network modeling of a complete ammonium nitrification system with parameter identification // Computer Research and Modeling, 2026, vol. 18, no. 4, pp. 909-928
Citation in English: Grechneva A.O., Tarasov V.D., Tarkhov D.A., Shatrov A.V. Neural network modeling of a complete ammonium nitrification system with parameter identification // Computer Research and Modeling, 2026, vol. 18, no. 4, pp. 909-928
DOI: 10.20537/2076-7633-2026-18-4-909-928
URL: http://crm-en.ics.org.ru/journal/article/3752/
Creative Commons License This work is licensed under a Creative Commons Attribution-NoDerivs 3.0 Unported License.

Copyright © 2026 Grechneva A.O., Tarasov V.D., Tarkhov D.A., Shatrov A.V.

Indexed in Scopus

Full-text version of the journal is also available on the web site of the scientific electronic library eLIBRARY.RU

The journal is included in the Russian Science Citation Index

The journal is included in the RSCI

International Interdisciplinary Conference "Mathematics. Computing. Education"

Website Copyright © 2009–2026 Institute of Computer Science