Document Type : Original research
School of Chemical and Energy Engineering, Faculty of Engineering, Universiti Teknologi Malaysia, 81310, Johor, Malaysia
Manufacturing Division, Pengerang Refining Company Sdn Bhd, PICMO B2, Pengerang Integrated Complex, 81600 Johor, Malaysia
Group Technical Solutions, Petroliam Nasional Berhad (PETRONAS), The Intermark Tower, 55000 Kuala Lumpur, Malaysia
The study was conducted in the actual world-scale olefin plant with a focus on measuring the impact of identified controlled variables at the steam cracker furnace towards the propylene yield. Surface response analysis was conducted in the Minitab software version 20 using the historical data after the clearance of both the outliers and residuals to ensure the analysis was conducted as normal data. Surface response analysis is a robust mathematical and statistical approach that is having a good potential to be systematically utilized in the actual large-scale olefin plant as an alternative to the expensive olefin simulation software for process monitoring. The analysis was conducted to forecast the maximum propylene yield in the studied plant with careful consideration to select only significant variables, represented by a variance inflation factor (VIF) <10 and p-value <0.05 in the analysis of variance (ANOVA) table. The final model successfully concluded that propylene yield in the studied plant was contributed by the factors of 0.00496, 0.00204, and -3.96 of hearth burner flow, dilution steam flow, and naphtha feed flow respectively. The response optimizer also suggested that the propylene yield from naphtha pyrolysis cracking in the studied plant could be maximized at 11.47% with the control setting at 10,004.36 kg/hr of hearth burner flow, 40,960 kg/hr of dilution steam flow, and 63.50 t/hr of naphtha feed flow.