The paper addresses the optimisation of control strategies for biological wastewater treatment in a sequencing batch reactor (SBR) system. The classical control structure was extended by introducing a supervisory control layer, within which a two-stage optimisation framework was designed and implemented, consisting of an outer and an inner optimiser. The main objective was to determine the optimal operating parameters of the reactor cycle, including the number and duration of biological reaction phases as well as the nonlinear reference trajectory of dissolved oxygen (DO) concentration (DOref), in order to reduce energy consumption while ensuring the required quality of treated wastewater in accordance with the applicable water-law regulations. The study was carried out in the MATLAB environment using a simulation model developed on the basis of a real installation located in Swarzewo (Northern Poland). The optimisation problem was solved using stochastic methods, namely genetic algorithms (GA) and particle swarm optimisation (PSO). The simulation results confirmed the effectiveness of the proposed approach and showed that the application of the supervisory two-stage optimisation layer improves the operational performance of the wastewater treatment plant. Reductions in chemical oxygen demand (COD), total nitrogen (TN), and total phosphorus (TP) of 100%, 71.6%, and 84.1%, respectively, were achieved.
Hydrotreating of bitumen-derived gas oils is essential for producing clean fuels, but the process is challenged by fines deposition, which leads to progressive pressure drop buildup and premature reactor shutdown. Pressure drop correlations can serve as valuable tools for anticipating such buildups. However, existing correlations, including the Ergun equation and its two-phase extensions, do not account for the hydrodynamic effects of fines deposition, thereby limiting their applicability. In this study, an Ergun-based model was developed to predict pressure drop during fines deposition. The model integrates a two-phase multiplier with a time-dependent fines deposition multiplier, both expressed as functions of operating conditions and packing properties. Model coefficients were estimated via non-linear regression using experimental data spanning temperatures of 350-390 °C, gas velocities of 0.03-0.10m/s, packing sizes of 2.1-2.7mm and fines sizes of 0.2-20µm. The integrated model demonstrated strong predictive capability across the examined range of conditions. Sensitivity analysis revealed that fines size had the strongest individual influence on pressure drop buildup, whereas gas velocity and temperature dominated when interaction effects were considered. The resulting framework provides a predictive tool for estimating pressure drop during fines deposition and offers valuable insights for mitigation strategies when processing fines-laden oil feeds.
