Perancangan Model Predictive Control pada Kolom Distilasi Biner Ammonia Stripper

Rahayu, Frely Novianti (2017) Perancangan Model Predictive Control pada Kolom Distilasi Biner Ammonia Stripper. Undergraduate thesis, Institut Teknologi Sepuluh Nopember.

[img]
Preview
Text
2413100105-Undergraduate_Theses.pdf - Published Version

Download (2MB) | Preview

Abstract

Ammonia stripper memisahkan aliran feed, aqueous ammonia kedalam dua produk, ammonia (distillate) dan air (bottom product). Untuk memperoleh produk distilasi dengan tingkat kemurnian tinggi sangat sulit jika menggunakan kontroler konvensional untuk distilasi karena sifat kolom yang sangat non-linear dan disebabkan oleh interaksi alami antara loop yang dikontrol. Tugas akhir ini dilakukan perancangan model predictive control yang mampu mengontrol sistem MIMO yang memiliki interaksi loop serta memiliki sifat non-linear. Langkah awal yang dilakukan adalah melakukan pemodelan dari proses yang terjadi pada ammonia stripper. Selanjutnya melakukan perancangan model predictive control yang dilanjutkan dengan tuning MPC dengan terlebih dulu melakukan pemodelan empirik FOPDT. Langkah terakhir adalah melakukan uji performansi pada MPC dengan memberikan uji disturbance, uji setpoint tracking dan melakukan perbandingan performansi kontrol antara pengendali PI dan MPC. Dari hasil perbandingan performansi kontrol, MPC memiliki hasil lebih baik daripada PI yaitu dengan nilai maximum overshoot sebesar 1,38%, settling time 3 detik dan error steasy state 2% sebesar 0,6495%. ============================================================ Ammonia stripper separates feed stream, aqueous ammonia into two products, ammonia (distillate) and water (bottom product). It is difficult to obtain high purity distillation product due to the natural interaction between the controlled loops. In this final project is designed model predictive control that able to control MIMO system which has loop interaction and non-linear characteristic. At first got the mathematical model of the process in ammonia stripper. Then designed model predictive control and followed by MPC tuning by first got empirical model of FOPDT. The final step is doing performance test on MPC byb giving disturbance test, setpoint tracking test and doing comparison of control performance between PI and MPC. The result is MPC has better result than PI with maximum overshoot 1,38%, settling time 3 seconds and 2% steady state error 0,6495%.

Item Type: Thesis (Undergraduate)
Uncontrolled Keywords: model predictive control, tuning MPC, kolom distilasi, ammonia stripper
Subjects: T Technology > TK Electrical engineering. Electronics Nuclear engineering
T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK5105.546 Computer algorithms
Divisions: Faculty of Industrial Technology > Physics Engineering > (S1) Undergraduate Theses
Depositing User: Rahayu Frely Novianti
Date Deposited: 12 Oct 2017 04:40
Last Modified: 08 Mar 2019 02:17
URI: http://repository.its.ac.id/id/eprint/46574

Actions (login required)

View Item View Item