Implementasi Model Predictive Control (MPC) Pada Kestabilan Sistem Kendali Tegangan DC Untuk Beban Dinamis Di Modul Remote Laboratory

Aritonang, Mario (2025) Implementasi Model Predictive Control (MPC) Pada Kestabilan Sistem Kendali Tegangan DC Untuk Beban Dinamis Di Modul Remote Laboratory. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 2040211102-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (3MB) | Request a copy

Abstract

Sistem kendali generator DC di Remote Labotaroy saat ini menggunakan metode PID dan Fuzzy Logic Controller (FLC) yang memiliki keterbatasan dalam menjaga kestabilan tegangan dari keluaran generator DC. Hal ini terjadi ketika beban mengalami perubahan secara dinamis. Penelitian ini mengusulkan implementasi Model Predictive Control (MPC) untuk metode sistem kontrol yang dapat digunakan pada beban dinamis. Sistem memiliki komponen utama motor induksi AC tiga fasa (0,25HP), generator DC, sensor tegangan, Arduino Uno R3, Inverter Schneider ATV12. Algoritma akan diimplementasikan melalui software Simulink MATLAB untuk mengatur persamaan matematika. Pengujian pada sistem tanpa beban, beban minimal, dan beban meningkat yang telah dilakukan pada penelitian ini dengan menggunakan beberapa variabel parameter, menunjukan bahwa kontroler MPC mampu mencapai setpoint 5V dan memiliki nilai steady state error sebesar 0.1V pada parameter kontroler MPC paling optimal. Di tahap pengujian dengan variasi beban meningkat dan menurun, kontroler MPC mampu menstabilkan sistem dengan waktu untuk menstabilkan sistem selama ≈5 detik. Hasil pengujian menunjukan bahwa dampak dari penambahan nilai horizon prediksi (Np) dapat mempercepat nilai risetime sistem dan dampak dari memperkecil nilai bobot (λ) dapat mempercepat sistem untuk stabil dengan nilai setpoint, namun memiliki kekurangan menghasilkan nilai sinyal kontrol yang cukup besar di awal sistem.
=========================================================================================================================================
The DC generator control system at Remote Labotaroy currently uses the PID and Fuzzy Logic Controller (FLC) methods which have limitations in maintaining voltage stability from the DC generator output. This occurs when the load changes dynamically. This study proposes the implementation of Model Predictive Control (MPC) for a control system method that can be used on dynamic loads. The system has the main components of a three-phase AC induction motor (0.25HP), a DC generator, a voltage sensor, an Arduino Uno R3, and a Schneider ATV12 inverter. The algorithm will be implemented through MATLAB Simulink software to organize mathematical equations. Tests on the system without load, minimum load, and increased load that have been carried out in this study using several parameter variables, show that the MPC controller is able to reach a setpoint of 5V and has a steady state error value of 0.1V at the most optimal MPC controller parameters. In the testing phase with increasing and decreasing load variations, the MPC controller is able to stabilize the system with a time to stabilize the system for ≈5 seconds. The test results show that the impact of increasing the prediction horizon value (Np) can accelerate the system rise time value and the impact of reducing the weight value (λ) can accelerate the system to stabilize with the setpoint value, but has the disadvantage of producing a fairly large control signal value at the start of the system.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Model Predictive Control (MPC), Motor Induksi Tiga Fasa, Remote Laboratory, Model Predictive Control (MPC), Three-phase Induction Motor, Remote Laboratory
Subjects:	T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK2785 Electric motors, Induction.
Divisions:	Faculty of Vocational > 36304-Automation Electronic Engineering
Depositing User:	Mario Aritonang
Date Deposited:	04 Aug 2025 09:42
Last Modified:	04 Aug 2025 09:42
URI:	http://repository.its.ac.id/id/eprint/126442

Actions (login required)

View Item