Assessing and simulation of membrane technology for modifying starchy wastewater treatment

Abstract

In this study, a hydrophilic polyethersulfone membrane was used to modify the expensive and low efficient conventional treatment method of wheat starch production that would result in a cleaner starch production process. To achieve a cleaner production, the efficiency of starch production was enhanced and the organic loading rate of wastewater that was discharged into treatment system was decreased, simultaneously. To investigate the membrane performance, the dependency of rejection factor and permeate flux on operative parameters such as temperature, flow rate, concentration, and pH of feed were studied. Response surface methodology (RSM) has been applied to arrange the experimental layout which reduced the number of experiments and also the interactions between the parameters were considered. The maximum achieved rejection factor and permeate flux were 97.5% and 2.42 L min -1 m -2 , respectively. Furthermore, a fuzzy inference system was selected to model the nonlinear relations between input and output variable which cannot easily explained by physical models. The best agreement between the experimental and predicted data for permeate flux was denoted by correlation coefficient index (R 2 ) of 0.9752 and mean square error (MSE) of 0.0072 where defuzzification operator was center of rotation (centroid). Similarly, the maximum R 2 for rejection factor was 0.9711 where the defuzzification operator was mean of maxima (mom).

A Comprehensive Review on Membrane Fouling: Mathematical Modelling, Prediction, Diagnosis, and Mitigation

“…It was concluded that higher flux rates were predicted using the novel model, in contrast to estimations generated using the backpropagation (BP) neural network model. Another way AI can be utilized in membrane fouling management was introduced by Moghaddam et al [31], who used fuzzy logic (FL) to process variables. The relationship between the wastewater feed operating parameters (i.e., temperature, pH, and flow rate) and the membrane's fouling response was investigated.…”

Prediction of COD in industrial wastewater treatment plant using an artificial neural network

“…There are many statistical and AI-based predictive models used for this purpose. Among the widely and successfully used methods are various Artificial neural networks (ANNs) 5 , response surface methodology (RSM) 7 , 8 , radial basis function (RBF) 8 , and fuzzy logic 9 .…”

“…There are many statistical and AI-based predictive models used for this purpose. Among the widely and successfully used methods are various Artificial neural networks (ANNs) 5 , response surface methodology (RSM) 7,8 , radial basis function (RBF) 8 , and fuzzy logic 9 .ANNs, as computational techniques, are nonlinear models designed to mimic the functionality and decisionmaking processes of the human brain 10 . ANNs have been increasingly applied in various environmental modeling studies 11,12 and investigations into water quality issues 13,14 .…”

Artificial neural network modeling and optimization of wheat starch suspension microfiltration using twisted tape as a turbulence promoter

“…Therefore, the modeling approach including the direct analysis of experimental data can serve as an adequate alternative to models based on phenomenological hypotheses (knowledge-based models) (Hilal et al, 2008;Sahoo & Ray, 2006). Artificial neural network (ANN), an effective predictive model for nonlinear systems where mathematical models are not suitable, has been successfully employed to presume different aspects of membrane performances (Chellam, 2005;Guadix, Zapata, Almecija, & Guadix, 2010;Hedayati Moghaddam, Hazrati, Sargolzaei, & Shayegan, 2017;Liu et al, 2014). The advantage of empirical modeling tools in regard to theoretical models is reflected in the potentially rapid development of the objective function appropriate for the process optimization (Khayet, Cojocaru, & Essalhi, 2011).…”