ABSTRACT
Peristaltic pumping is used in membrane applications where high and sterile sealing is required. However, control is difficult due to the pulsating pump characteristics and the time-varying properties of the system. In this work, three artificial intelligence control strategies (artificial neural networks (ANN), fuzzy logic expert systems, and fuzzy-integrated local models) were used to regulate transmembrane pressure and crossflow velocity in a microfiltration system under high fouling conditions. A pilot plant was used to obtain the necessary data to identify the AI models and to test the controllers. Humic acid was employed as a foulant, and cleaning-in-place with NaOH was used to restore the membrane state. Several starting operating points were studied and setpoint changes were performed to study the plant dynamics under different control strategies. The results showed that the control approaches were able to control the membrane system, but significant differences in the dynamics were observed. The ANN control was able to achieve the specifications but showed poor dynamics. Expert control was fast but showed problems in different working areas. Local models required less data than ANN, achieving high accuracy and robustness. Therefore, the technique to be used will depend on the available information and the application dynamics requirements.
ABSTRACT
Optimization of the ultrafiltration (UF) process to remove colloidal substances from a paper mill's treated effluent was investigated in this study. The effects of four operating parameters in a UF system (transmembrane pressure (TMP), cross-flow velocity (CFV), temperature and molecular weight cut-off (MWCO)) on the average permeate flux (Jv), organic matter chemical oxygen demand (COD) rejection rate and the cumulative flux decline (SFD), was investigated by robust experimental design using the Taguchi method. Analysis of variance (ANOVA) for an L9 orthogonal array were used to determine the significance of the individual factors, that is to say, to determine which factor has more and which less influence over the UF response variables. Analysis of the percentage contribution (P%) indicated that the TMP and MWCO have the greatest contribution to the average permeate flux and SFD. In the case of the COD rejection rate, the results showed that MWCO has the highest contribution followed by CFV. The Taguchi method and the utility concept were employed to optimize the multiple response variables. The optimal conditions were found to be 2.0 bar of transmembrane pressure, 1.041 m/s of the cross-flow velocity, 15 °C of the temperature, and 100 kDa MWCO. The validation experiments under the optimal conditions achieved Jv, COD rejection rate and SFD results of 81.15 L·m-2·h-1, 43.90% and 6.01, respectively. Additionally, SST and turbidity decreased by about 99% and 99.5%, respectively, and reduction in particle size from around 458-1281 nm to 12.71-24.36 nm was achieved. The field-emission scanning electron microscopy images under optimal conditions showed that membrane fouling takes place at the highest rate in the first 30 min of UF. The results demonstrate the validity of the approach of using the Taguchi method and utility concept to obtain the optimal membrane conditions for the wastewater treatment using a reduced number of experiments.
