Process Parameters Optimization in Membrane Fabrication for Produced Water Treatment Using Response Surface Methodology (RSM) and Central Composite Design (CCD)

Abstract

Produced water is wastewater from oil production that must be treated well. Membrane is one alternative to water treatment technology based on filtration method. But, in the use of a membrane, there’s no exact variable optimal that influences performance of the membrane. This underlying research to assess factors that influences the performance of membrane to be more optimal. Therefore, the objectives of this study determine the optimum variable through Respond Surface Methodology and Central Composite Design. After getting the optimal condition then will check the stability of the membrane. This experiment of optimization of produced water with asymmetric membrane's Polyether sulfone (PES) using Response Surface is done with varying the Zeolite concentration by low level 1% weight and 3% weight, length of UV irradiation time low level 2 minutes and high level 6 minutes, thermal annealing low level 160 ºC and high-level 180ºC. An analyzer done in this research was by processing data research to make table and charts of the relationship between the result of this experiment with changed variable, namely variation of PES concentration, time of UV ray and thermal annealing by using Response Surface Methodology (RSM) and Central Composite Design (CCD).

Optimization of Crystallization Process Condition of Nutmeg Seed Oleoresin

“…Center point (0) was repeated for 3 times in 2 batch, so resulting 17 runs in each batch. The runs come from 2k+2k+no formula, where 'k' was the amount of independent variable and 'no' indicates the amount of repetition in every analysis [12]. Analysis was continued with analysis of variance to find out the independent variable that affects to the response significantly, fitted surface to find out 3-dimensions surface graphic of response to independent variable, Pareto chart to find out the effect of independent variable which either positive or negative to the response, critical values to find out independent variable of optimum response, and regression analysis to find out prediction of response equation mathematically.…”

Optimization of Chromium (VI) Adsorption using Microalgae Biomass (Spirulina sp.) and its Application in Leather Tannery Waste

“…A model is considered good when it has significant values for the response and insignificant values for the lack of fit test. An insignificant lack of fit test value shows that the model is suitable or sufficiently represents the data [13]. The results obtained from 16 experiments are presented in Table 4.…”

“…Optimization aims to develop an empirical model to find optimal parameters for maximum or minimum responses [13]. This step can be performed using the RSM method [14].…”

“…The accuracy test of the regression model was carried out by dividing the mean of squares of regression (MSreg) and the mean of squares of residuals (MSres). Both metrics were then compared with the table value of Fisher (F), where both MSreg and MSres represented the Sum of Squares divided by the Degree of Freedom [13]. The results of the subsequent variance analysis followed the hypotheses: H0 = all βi values are zero H1= at least one βi is not zero H0 is declared true if Fcount<FTable.…”

“…Another test was conducted to strengthen the assumption of model suitability through a lack of fit test [26], following the hypotheses: H0 = there is no lack of fit; the model created is suitable for the data H1 = there is a lack of fit; the model created does not represent the data H0 was accepted when the p-value was more than the degree of significance, α = 0.05 [13]. The accuracy of this model was also reflected in the value of the coefficient of determination R 2 .…”