Related Experiment Video
Updated: Dec 13, 2025
06:37
Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention
Published on: December 15, 2023
5.0K
Prediction the performance of multistage moving bed biological process using artificial neural network (ANN)
1Department of Chemical Engineering, College of Engineering, Qatar University, P. O. Box 2713, Doha, Qatar.
The Science of the Total Environment
|July 30, 2020
Summary
An artificial neural network (ANN) effectively models and controls enhanced nutrient removal biological processes (ENR-BP) for wastewater treatment. This approach achieves high removal rates for COD, NH4+, and TP, improving process efficiency and control.
Area of Science:
- Environmental Engineering
- Biotechnology
- Artificial Intelligence in Environmental Science
Background:
- Biological nutrient removal processes (BPs) in wastewater treatment are complex and difficult to model due to inherent uncertainties and dynamic nature.
- Traditional methods often rely on approximations, limiting effective process control and optimization.
- Need for advanced modeling techniques to accurately predict and manage enhanced nutrient removal biological processes (ENR-BP).
Purpose of the Study:
- To develop and validate an artificial neural network (ANN) algorithm for simulating, modeling, and controlling a three-stage ENR-BP.
- To evaluate the impact of key operational parameters (SALR, OMs, nutrients, Qfeed, HRT, IRF) on ENR-BP performance.
- To achieve stringent wastewater discharge limitations through optimized biological nutrient removal.
Main Methods:
- Development of a three-stage ENR-BP incorporating anaerobic/anoxic stages and a moving bed biofilm reactor (MBBR).
- Utilized experimental data for iterative training and testing to establish the optimal ANN architecture.
- Investigated the influence of surface area loading rate (SALR), organic matter (OMs), nutrients (N & P), feed flow rate (Qfeed), hydraulic retention time (HRT), and internal recycle flow (IRF).
Main Results:
- Achieved high removal efficiencies: COD (89.2–98.3%), NH4+ (88.5–98.9%), and TP (77.9–99.9%).
- Optimal performance achieved at a total HRT of 13.3 h and an IRF of 1.75.
- ENR-BP demonstrated four-fold higher removal efficiencies than suspended growth processes and comparable to 5-stage Bardenpho-MBBR.
Conclusions:
- The developed ANN model provides a robust and efficient tool for predicting and forecasting ENR-BP performance.
- The ENR-BP mechanism effectively utilizes OMs for phosphorus bio-uptake and simultaneous nitrification and denitrification (SND).
- ANN-based modeling enhances the control and operational effectiveness of complex biological wastewater treatment processes.
