Determine the most effective process control parameters on activated sludge based on particle swarm optimisation algorithm (Case Study: South wastewater treatment plant of Tehran)
Mina Rafati,Mohammad Pazouki,Hossein Ghadamian,Azarmidokht Hossein nia,Ali Jalilzadeh
DOI: https://doi.org/10.1080/03067319.2020.1836169
2021-01-12
International Journal of Environmental Analytical Chemistry
Abstract:Wastewater treatment plants (WWTPs) are major energy consumers and cause environmental impact on receiving waters. Many WWTPs are operated in a less-than- optimal manner with respect to both treatment and energy efficiency. A better solution is to optimize the operation and processes of the existing WWTPs. Therefore, the most effective process parameter for providing control strategies was determined. The South WWTP of Tehran was designed and simulated based on the activated sludge model using MATLAB/ Simulink software for the first time in the country to obtain the factors affecting WWTP. To calibrate this model some kinetic and stoichiometric coefficients were determined. Then the simulator validation was performed by three error calculation methods. Finally, the values of three main controllers, including the optimum rate or return activated sludge (RAS), internal recycle (IR) and oxygen transfer coefficient (KLa) are determined using the Particle Swarm Optimization algorithm. According to the results, coefficients such as Y, kd, K and KS for Oxidation-Ditch process were in the range of 0.303-0.331mgVSS /mg sCOD, 0.030-0.033 1/day, 1.65-1.93 1/day and 37.6-44.92 mg sCOD/l, respectively and the Mean of COD, BOD5, TSS and TN removal was obtained 94.8 ± 0.4, 97.3 ± 0.65, 94.7 ± 1.5 and 56 ± 7.46, respectively. Also, the percentage of Root Mean Square for TN, TSS, COD, BOD5 was 3.14%, 2.95%, 3.13% and 5.2%, respectively, Pearson correlation coefficient was 0.88, 0.93, 0.89, 0.99 and absolute mean error of 2.62%, 2.47%, 2.6% and 3.58%, respectively, which shows that the simulation outputs are compatible with the effluent of plant. To achieve the best process performance conditions, RAS and IR must be considered for 1.7 and 0.8 percent of influent wastewater and KLa is suggested to be 154 d-1. Therefore, by using the optimization performed, the effect of other controllers on the process can be investigated and selected.
chemistry, analytical,environmental sciences