Kontrol Temperature Pada Continuous Stirred Tank Reactor (CSTR) Menggunakan Neural Network Model Predictive Control (NNMPC)

Armananta, Muhammad Loudy (2024) Kontrol Temperature Pada Continuous Stirred Tank Reactor (CSTR) Menggunakan Neural Network Model Predictive Control (NNMPC). Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5022201226-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 October 2026. Download (3MB) | Request a copy

Abstract

Continuous Stirred Tank Reactor (CSTR) merupakan sistem yang umum digunakan dalam industri kimia untuk proses pencampuran dan reaksi kimia. Kontrol yang efektif terhadap CSTR sangat penting untuk memastikan kualitas produk, efisiensi proses, dan keamanan operasi. Proses ini merupakan sistem nonlinear, dikarenakan sistemnya sangat tidak proporsional karena reaksi didalamnya selalu berubah ubah seperti temperatur, konsentrasi produk, dan lainnya sehingga kontroler konvensional tidak cukup sensitif dalam merespon perubahan setpoint yang terjadi. Pada penelitian ini akan dilakukan pemodelan sistem Continuous Stirred Tank Reactor (CSTR) untuk mengkontrol temperatur reaktor (T). Pemodelan sistem tersebut digunakan untuk merancang neural network model predictive control (NNMPC) untuk memprediksi kinerja dan performansi sistem kendali yang ideal karena kelebihannya dalam melakukan pembelajaran dari data input dan output suatu proses. Hasil penelitian sistem kontroler NNMPC menunjukkan performa yang
lebih baik dibandingkan sistem kontroler konvensional ditunjukkan oleh nilai yang lebih kecil pada parameter settlingtime sebesar 130 detik, dan Integral Absolute Error (IAE) sebesar 5,512558.
============================================================
=======================================================================
Continuous Stirred Tank Reactor (CSTR) is a system commonly used in the chemical industry for mixing and chemical reaction processes. Effective control of CSTR is essential to ensure product quality, process efficiency and operational safety. This process is a nonlinear system, because the system is very disproportionate because the reactions in it are always changing, such as temperature, product concentration, etc. so that conventional control is not sensitive enough to respond to setpoint changes that occur. In this research, the Continuous Stirred Tank Reactor (CSTR) system will be modeled to control the reactor temperature (T). This system modeling is used to design neural network model predictive control (NNMPC) to predict the performance and performance of an ideal control system because of its advantages in learning from the input and output data of a process. The results of the NNMPC controller system research show better performance than the conventional controller system indicated by a smaller value in the settlingtime parameter of 130 seconds, and Integral Absolute Error (IAE) of 5.512558.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Continuous Stirred Tank Reactor (CSTR), neural network model predictive control (NNMPC), Integral Absolute Error (IAE), Kontrol Temperatur, Continuous Stirred Tank Reactor (CSTR), neural network model predictive control (NNMPC), Integral Absolute Error (IAE), Temperature Control.
Subjects:	Q Science > QA Mathematics > QA76.87 Neural networks (Computer Science) T Technology > TJ Mechanical engineering and machinery > TJ217.6 Predictive Control
Divisions:	Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Electrical Engineering > 20201-(S1) Undergraduate Thesis
Depositing User:	Muhammad Loudy Armananta
Date Deposited:	31 Jul 2024 01:54
Last Modified:	31 Jul 2024 01:54
URI:	http://repository.its.ac.id/id/eprint/110984

Actions (login required)

View Item