Treffer: Modeling of Wastewater Treatment and Dynamic Optimization Control of Dissolved Oxygen Concentration in Aeration Tanks.

The wastewater biochemical treatment process exhibits significant nonlinearity, posing challenges for the design and optimization of control strategies. To address these challenges, Benchmark Simulation Model No. 1 for activated sludge wastewater treatment was first developed using MATLAB's S-function and the Simulink platform. Comprehensive open-loop simulations were conducted using both constant and dynamic influent data under different weather conditions (clear, rainy, and stormy). Compared to actual reference data, the simulations yielded a mean absolute error of only 0.051152 g/m3 in the effluent TSS concentration of the top layer of the secondary clarifier, verifying the accuracy and effectiveness of the model. Second, PI control experiments on dissolved oxygen concentration revealed that, under stormy conditions, the total nitrogen content in the wastewater failed to meet regulatory standards. Nonlinear model predictive control was applied to address this, and random excitation was introduced into the model's open-loop simulations to enhance prediction accuracy. The resulting data were used to optimize and train various machine learning models, including backpropagation (BP) neural networks, random forests, support vector machines, and least squares support vector machines (LSSVMs), using the particle swarm optimization (PSO) algorithm. Third, the PSO-LSSVM prediction model, which demonstrated the highest prediction accuracy, was selected and combined with the fruit fly optimization algorithm to determine the optimal control parameters for model predictive control. Simulation results confirmed that all performance indicators met the standards under all tested weather conditions. Finally, traditional constant dissolved oxygen control strategies have been shown to result in energy waste. By dynamically adjusting the dissolved oxygen set point, aeration energy consumption was reduced by 3.27% without significantly affecting effluent water quality. [ABSTRACT FROM AUTHOR]

Copyright of Journal of Environmental Engineering is the property of American Society of Civil Engineers and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)