Statistical Process Control for Depropanizer Column at Petronas Gas Berhad, Kerteh

The final report is made in order to give all the details on the Final Year Project II
which is "Statistical Process Control for Depropanizer Column at Petronas Gas
Berhad (PGB), Kerteh ". This report is divided into five main chapters which are
Introduction, Literature Review, Methodology, Results & Discussions and Conclusion.
Statistical Process Control (SPC) has a very high demand in industry right now as
it provides a way to monitor process behavior and able us to analyze the variations in the
process that may affect the quality of the end product. In this project, the focus in on the
depropanizer column at Petronas Gas Berhad (PGB), Kerteh. The short term targets are to
apply Statistical Process Control on the column stated, to measure and analyze the
variation in the processes, to monitor the consistency of processes used to manufacture
the product as designed and finally to suggest the best way in controlling all the variables
at the column. While for long term is to implement the results in our industry.
Many variables have to be considered in order to complete the project, which are:
(a) Calculated data surrounding the depropanizer column which include all the tag names,
(b) Tag names and description, (c) Description of the depropanizer column and (d) Flow
sheet for the column showing all the tag name surrounding the column. Mainly, two tools
are required in executing this project which are (a) SSPS software and (b) LIMS -
Laboratory Information Management System (in PGB, Kerteh). From here, an early
analysis on all the variables obtained by using Microsoft Excel has been made. Mainly,
the discussion is about the input variables that affecting the output variable, which in this
case the output variable is the C3 composition. Some problems have been identified
during the process of analyzing the data using Microsoft Excel.
Simulation using SPSS software has been completed which includes:
a) Descriptive Statistics a) Crosstabs
b) Histograms b) One Way Anova
c) Correlations c) Paired T -test
d) Scatter Plots d) Linear Regressions Analysis has been completed for the results obtained from SPSS, focusing on the
critical components inside the overhead product composition which is C3 and some input
data that will most probably affect the overhead product composition:
i) Reflux flow ( 4FC6203 .PV)
ii) Energy input inside the column ( 4T1623l.PV)
iii) Feed conditions (4FY62022.PV, 4TI6009.PV)
It is proven that these four main input variables have a very strong relationship
with C3 composition inside the overhead product and they all come from a general
population mean. Increasing or decreasing their values will give a great impact to the C3
composition. As the samples proved to come from a general population mean, an
optimum operating condition could be produced from the average data, in order to
maintain C3 composition within the desired value (98.48 mole%):
Input Description Optimum operating
variables conditions suggested
4FC6203.PV Reflux flow 112.96 m,/hr
4TI623l.PV Energy input inside the column 116.03 oc
I Reboiler temperature
4FY62022.PV Feed flowrate 143.36 m,/hr
4TI6009.PV Feed temperature 95.81 oc
However, these optimum operating conditions suggested must be checked again
so that it will not violate the design operating conditions.
This project will not only improve the existing process control of the column but
also improve the quality of end product and saving the cost to operate the column.