Darmawan, Rheza Qashmal (2016) Desain Autotuning Kontroler PID Berbasis Algoritma Neural-Network Untuk Sistem Pengaturan Cascade Level dan Flow Liquid Pada Plant Coupled Tanks. Undergraduate thesis, Institut Teknologi Sepuluh Nopember.
Preview |
Text
2212100160-Undergraduate-Theses.pdf Download (3MB) | Preview |
Abstract
Pada industri proses yang melibatkan fluida, suatu fluida akan
dipompa dan dialirkan dari satu tangki ke tangki yang lain untuk diolah.
Pemindahan cairan dari satu tangki ke tangki yang lain meyebabkan
berubahnya level fluida dalam tangki. Dalam pengaturan level,
pemindahan cairan biasa disebut sebagai pembebanan pada level.
Perubahan beban ini dapat mempengaruhi dari kinerja kontroler..
Kontroler yang banyak digunakan di industri proses adalah kontroler PID
karena kesederhanaan struktur dan kehandalannya. Pada penerapan
kontroler PID, tuning parameter kontroler sering dilakukan dengan
prosedur trial and error. Untuk tetap memenuhi spesifikasi kontrol yang
diharapkan, maka perlu dilakukan tuning ulang parameter kontroler PID.
Kontroler PID Neural Network ini didesain untuk dapat melakukan
autotuning pada parameter kontroler PID sehingga dapat mengatasi
perubahan parameter pada plant dan menjaga performa dari plant.
Berdasarkan hasil simulasi, sistem pengaturan level air pada plant coupled
tank dengan PID Neural Network lebih baik dengan nilai RMSE 0,044 %
daripada kontroler PID dengan nilai RMSE 0.35 %. Pada pengujian beban
kontroler PI dengan konfigurasi kontrol cascade mampu memberikan
hasil yang lebih baik dengan nilai RMSE 1.13 %.
=================================================================================
In the industrial processes involving fluid, a fluid to be pumped and
drained from one tank to another tank for processing. The transfer of
liquid from one tank to another led to changes in the liquid level in the
tank. In the level control, commonly referred to as the displacement fluid
loading level. This load changes may affect the performance of the
controller. The main control strategy used is based on the PID controller
design because is simple and robustness. In many times the controller
tuning is done by trial and error. Plant in the industry can change the
parameters that result from changes in the load on the plant. To keep
control of who is expected to meet the specifications it is necessary to retuning
PID controller parameters. Neural Network PID controller is
designed to perform autotuning in PID controller parameters so that it
can cope with changes in the parameters of the plant and maintain the
performance of the plant. Based on simulation results, level control system
on plant coupled tank with PID Neural Network is better with RMSE value
0.044% then controller PID with RMSE value 0.35%. For load
disturbance testing controller PID with cascade configuration control
able to profit to better result with RMSE value 1.13 %.
| Item Type: | Thesis (Undergraduate) |
|---|---|
| Additional Information: | RSE 629.831 2 Dar d |
| Uncontrolled Keywords: | Coupled tanks; kontroler PID; Neural Network; PID Controller |
| Subjects: | Q Science > QA Mathematics > QA76.87 Neural networks (Computer Science) T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK5105.546 Computer algorithms |
| Divisions: | Faculty of Industrial Technology > Electrical Engineering > 20201-(S1) Undergraduate Thesis |
| Depositing User: | Anis Wulandari |
| Date Deposited: | 09 Jun 2017 02:42 |
| Last Modified: | 26 Dec 2018 07:14 |
| URI: | http://repository.its.ac.id/id/eprint/41543 |
Actions (login required)
 |
View Item |