Perancangan Sistem Pengendalian Level Berbasis Model Reference Adaptive Control (MRAC) Pada Separator PT.Petrokimia Gresik

Widiyanti, Ira Nur (2017) Perancangan Sistem Pengendalian Level Berbasis Model Reference Adaptive Control (MRAC) Pada Separator PT.Petrokimia Gresik. Undergraduate thesis, Institut Teknologi Sepuluh Nopember.

[thumbnail of 2413100021-Undergraduate_Theses.pdf]
Preview
Text
2413100021-Undergraduate_Theses.pdf - Published Version

Download (2MB) | Preview

Abstract

Separator merupakan plant yang berfungsi untuk memekatkan larutan urea hingga konsentrasi 99,7 %. Pemekatan dilakukan dengan cara pemanasan pada Final Concentration dan pemvakuman di Separator. Waktu pemekatan dalam Separator diatur dengan ketinggian level bawah vessel. Level operasi dijaga pada 0,8 m . Pengendalian yang digunakan pada PT.Petrokimia Gresik adalah sistem pengendalian PI. Terdapat kelemahan pada sistem pengendalian PI yaitu ketika ada disturbance pada sistem, sistem harus dituning kembali satu-persatu setiap instrumen. Sehingga, pada tugas akhir ini dilakukan perancangan sistem pengendalian MRAC yang mampu beradaptasi ketika terjadi perubahan parameter plant seperti adanya gangguan. Langkah pertama yang dilakukan adalah pengambilan data dan spesifikasi separator,aktuator dan sensor. Langkah kedua adalah pemodelan separator, aktuator dan sensor. Langkah ketiga adalah perancangan MRAC (Model Reference Adaptive Control ) meliputi perancangan model reference,algoritma adjustment mechanism, dan perancangan PI.. Langkah terakhir adalah melakukan pengujian terhadap performansi sistem MRAC yang telah dirancang dengan pengujian variasi gain adaptasi, pengujian tracking set point naik dan turun serta pengujian pemberian gangguan. Nilai gain adaptasi yang sesuai untuk sistem MRAC adalah
훾 1= 0,2649 dan
훾 2= 0,0008. Perancangan kontroler
dan sistem MRAC pada separator
dapat beroperasi dengan optimal karena
sistem dapat beradaptasi terhadap adanya
perubahan set point naik dan turun serta dapat beradaptasi terhadap adanya gangguan dengan baik
========================================================================================================================
Separator is a plant that serves to concentrate urea solution to 99.7%
concentration. The concentration
made
by
warming up the Final
Concentration and vacuum in Separator. The concentration time in the
Separator is set to the level of the lower level of th
e vessel. Operating
level is maintained at 0.8 m. The control used in PT.
Petrokimia Gresik is
PI control
system. There is a weakness in the PI control system
,
when
there is disturbance on the system, the system must be redone one by one
each instrument. T
hus, in this final project design
ing
MRAC control
system that can adapt when there is a change of plant
parameters such as
disturbance
. The first step taken is data retrieval and separator
specifications, actuators and sensors. The second step is
separator,
actuator and sensor modeling. The third step is the design of MRAC
(Model Reference Adaptive Control) covering reference model design,
adjustment mechanism algorithm, and PI design. The final step is to test
the performance of MRAC system which
has been designed by testing
adaptation gain variation, testing point set point up and down As well as
testi
ng the provision of disturbance
. The corresponding adaptation gain
values
for the MRAC system are

1
= 0,2649 and

2
= 0,0008
.
The
design of
the controller and MRAC system on the separator can operate
optimally because the system can adapt to the change of set point up and
down and can adapt to the existence of the disturbance well.

Item Type: Thesis (Undergraduate)
Uncontrolled Keywords: Gain adaptasi, PI, Separator, Sistem MRAC
Subjects: T Technology > TJ Mechanical engineering and machinery > TJ217.6 Predictive Control
Divisions: Faculty of Industrial Technology > Physics Engineering > 30201-(S1) Undergraduate Thesis
Depositing User: Widiyanti Ira Nur
Date Deposited: 06 Sep 2017 06:46
Last Modified: 05 Mar 2019 07:05
URI: http://repository.its.ac.id/id/eprint/43420

Actions (login required)

View Item View Item