Pemodelan dan Kendali Optimal Pada Reaktor Fermentasi Nira Aren untuk Produksi Bioetanol

Zahirah, Aisyah Fikriyah (2026) Pemodelan dan Kendali Optimal Pada Reaktor Fermentasi Nira Aren untuk Produksi Bioetanol. Masters thesis, Institut Teknologi Sepuluh Nopember.

Text 6002241014-Master_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (1MB) | Request a copy

Abstract

Fermentasi bioetanol nira aren memiliki dinamika termal dan tekanan gas yang saling berinteraksi dan berubah seiring waktu. Suhu ekstrem berisiko mematikan ragi (Saccharomyces cerevisiae), sedangkan akumulasi karbon dioksida dapat memicu overpressure pada reaktor. Penelitian ini bertujuan untuk merancang dan mensimulasikan sistem Linear Time-Varying Model Predictive Control (LTV-MPC) pada bioreaktor guna mengendalikan suhu nira dan tekanan reaktor secara simultan. Metode ini dipilih secara khusus karena kemampuannya dalam memprediksi trayektori sistem guna meredam dinamika non-linear tanpa melanggar batasan keamanan alat. Model matematika diturunkan dari hukum kekekalan massa dan energi. Algoritma sistem dirancang dengan membatasi seberapa cepat pemanas dan katup gas boleh berubah. Pendekatan ini bertujuan untuk meredam sisa panas yang dapat membuat suhu melewati target batas aman serta mencegah pergerakan katup yang terlalu agresif. Perhitungan komputasi kemudian dilakukan untuk mencari langkah pengendalian yang paling efisien, dengan tetap menjadikan keselamatan operasional reaktor sebagai prioritas utama. Berdasarkan hasil simulasi, metode LTV-MPC terbukti andal dalam menjaga tekanan reaktor tetap stabil di batas aman 145 psi tanpa adanya lonjakan yang membahayakan. Keberhasilan menjaga tekanan ini tidak lepas dari kemampuan sistem dalam mengambil keputusan yang aman ketika dihadapkan pada situasi kritis. Sistem menunjukkan perilaku adaptif dengan membatasi pemanasan sebelum suhu mencapai batas aman 50 C, sehingga risiko kelebihan tekanan akibat pemuaian gas ideal dapat dicegah. Evaluasi pada kondisi suhu lingkungan konstan dan berfluktuasi menunjukkan bahwa LTV-MPC tetap andal (robust) dalam meminimalkan risiko terjadinya ledakan.
=================================================================================================================================
Palm sap bioethanol fermentation involves interacting thermal and gas-pressure dynamics that vary over time. Extreme temperatures may cuse the death of the yeast (Saccharomyces cerevisiae), while carbon dioxide accumulation can lead to reactor overpressure. This study aims to design and simulate a Linear Time-Varying Model Predictive Control (LTV-MPC) system for a bioreactor to simultaneously regulate the temperature of the palm sap and the reactor pressure. This method was specifically selected because of its ability to predict system trajectories and mitigate nonlinear dynamics without violating the equipment’s safety constraints. The mathematical model was derived from the laws of mass and energy conservation. The control algorithm was designed by limiting the rates at which the heater output and gas valve position could change. This approach was intended to reduce residual heat that could cause the temperature to exceed the safe target limit and to prevent excessively aggressive valve movements. Computational calculations were subsequently performed to determine the most efficient control actions while maintaining reactor operational safety as the primary priority. Based on the simulation results, the LTV-MPC method proved reliable in maintaining the reactor pressure at the safe limit of 145 psi without hazardous pressure spikes. This successful pressure regulation was supported by the system’s ability to make safe decisions under critical conditions. The system demonstrated adaptive behavior by limiting the heating input before the temperature reached the safe limit of 50◦C, thereby preventing the risk of excessive pressure caused by ideal-gas expansion. Evaluations under both constant and fluctuating ambient-temperature conditions showed that the LTV-MPC remained robust in minimizing the risk of explosio

Item Type:	Thesis (Masters)
Uncontrolled Keywords:	Kendali Multi-Input; Pemodelan Matematika; Linear Time-Varying Model Predictive Control} (LTV-MPC); Fermentasi Bioetanol; Back-Pressure
Subjects:	Q Science > QA Mathematics > QA401 Mathematical models. Q Science > QA Mathematics > QA402 System analysis.
Divisions:	Faculty of Science and Data Analytics (SCIENTICS) > Mathematics > 44101-(S2) Master Thesis
Depositing User:	Aisyah Fikriyah Zahirah
Date Deposited:	29 Jul 2026 01:52
Last Modified:	29 Jul 2026 01:52
URI:	http://repository.its.ac.id/id/eprint/139043

Actions (login required)

View Item