    |
Objective & Scope
1. Modeling of SBR process 2. Real-time control of SBR process 3. Hybrid neural network approaches in the modeling of a full-scale process Main Contents With increasingly stringent effluent requirement and loads on existing plants, the development of efficient operating strategies for wastewater treatment plants has become a very important problem. The improvement of the process efficiency largely depend on a better understanding of process behavior, the amount and quality of information that is available about the system, and the ability to simulate and predict the process dynamics. The principal goal of the current study is to carry out a comprehensive process modeling and control that aim to improve the operation of wastewater treatment. Especially, neural network techniques were used for the modeling of biological nutrient removal systems. In monitoring and controlling wastewater treatment processes, on-line information of nutrient dynamics is very important. However, these variables are determined with a significant time delay. Although the final effluent quality can be analyzed after this delay, it is often too late to make proper adjustments. Therefore, a neural network approach, a software sensor, was proposed to overcome this problem. A bench-scale sequencing batch reactor (SBR) used for advanced wastewater treatment (BOD plus nutrient removal) was employed to develop the neural network model. In order to improve the network performance, the structure of neural network was arranged in such a way of reflecting the change of operational conditions within a cycle. Real time estimation of phosphate, nitrate, and ammonium concentrations was successfully carried out with the on-line information of the SBR system only. Simultaneous biological nutrient removal enhancing phosphate uptake under anoxic conditions were investigated in an anaerobic-aerobic-anoxic-aerobic sequencing batch reactor((AO)2 SBR). The system showed stable phosphorus and nitrogen removal. Also the pH, ORP profiles were used to establish the real-time control strategy to determine the duration of operational phase of the (AO)2 SBR. Hybrid neural network approaches were applied in the modeling of a full-scale industrial wastewater treatment process. Initially, process data analysis was performed upon actual operational data using Principal Component Analysis. Secondly, a simplified mechanistic model and neural network model were developed based on specific mechanistic model and neural network model were developed based on specific process knowledge and operational data of the coke wastewater treatment process, respectively. Finally, a neural network was incorporated into the mechanistic model in both parallel and serial configurations. Simulation results indicate that the parallel hybrid neural modeling approach is a useful tool for the accurate and cost-effective modeling of wastewater treatment processes, in the absence of a reasonably accurate process model. References 1) Neural network modeling of biological wastewater treatment processes Dae Sung Lee, Thesis of ph. D.(2000), POSTECH, KOREA 2) Neural network modeling for on-line estimation of nutrient dynamics in sequentially-operated batch reactor D.S. Lee and J.M. Park, J. Biotechnol., 75, 229 (1999) 3) Biological nitrogen removal with enhanced phosphate uptake in a sequencing batch reactor using single sludge system D.S. Lee, C.O. Jeon, and J.M. Park, Water Research, 35(16), 206 (2001) 4) Hybrid neural network modeling of a full-scale wastewater treatment process D.S. Lee, C.O. Jeon, J.M. Park, and K.S. Chang, Biotechnol. Bioeng., 78(6), 670 (2002) 5) List based threshold accepting algorithm for zero-wait scheduling of multiproduct batch plants D.S. Lee, V.S. Vassiladis, and J.M. Park, Ind. Eng. Chem. Res., 41(25), 6579 (2002) |
(If you find any false information hear, please contact webmaster!)