Modeling of preparation conditions of PES ultrafiltration hollow fiber membranes using statistical regression techniques

Abstract

Mathematical modeling of the spinning process is crucial for a better understanding of the process variables and process functionality in membrane development. Due to the broad use and key importance of mathematical models in chemical process engineering, experimental design is becoming essential for the rapid development and validation of these empirical models. This work used the design of experiment methodology and aimed to predict the performance of ultrafiltration systems for water treatment by considering the statistical regression technique as an important approach for modeling flux. The utilization of regression modeling was also explored to show the principle elements for predicting flux in the spinning process. In order to investigate how proficient the statistical regression technique is at approximating the predicted value for flux, a real spinning experiment was conducted in this study. In this experiment, 30 samples of data were collected based on a half fractional factorial experiment with design resolution V, as well as 4 replications of center points and 10 axial points. The spinning factors that were investigated are the dope extrusion rate, air gap length, coagulation bath temperature, bore fluid ratio, and post-treatment time for predicting the corresponding flux. The regression model obtained shows that there is a correlation between the experimental data and predicted values. The results of the proposed model can be used to give a good prediction of the spinning process during membrane fabrication.