Pengembagan Sistem Kontrol Kecepatan Pada Motor BLDC Berbasis FOC Menggunakan Metode PI dan Reinforcement Learning

Amrulloh, Prishandy Hamami (2026) Pengembagan Sistem Kontrol Kecepatan Pada Motor BLDC Berbasis FOC Menggunakan Metode PI dan Reinforcement Learning. Masters thesis, Institut Teknologi Sepuluh Nopember.

Text 6022232003-Master_Thesis.pdf Restricted to Repository staff only Download (10MB) | Request a copy

Abstract

Motor BLDC banyak digunakan dalam aplikasi industri, termasuk kendaraan listrik dan robotika. Hal ini dikarenakan motor BLDC memiliki efisiensi energi, keandalan, dan kinerja tinggi. Pengendalian motor BLDC dalam kondisi dinamis yang melibatkan perubahan kecepatan dan beban sering kali menghadapi tantangan, terutama ketika menggunakan kontrol PI konvensional yang memerlukan penyetelan manual yang rumit. Tesis ini akan mengembangkan sistem kontrol untuk motor Brushless DC (BLDC) berbasis Field-Oriented Control (FOC) menggunakan metode Proportional-Integral (PI) yang dioptimalkan dengan Reinforcement Learning (RL). Tesis ini mengusulkan penggunaan RL, khususnya algoritma Deep Deterministic Policy Gradient (DDPG), untuk mengoptimalkan kontrol PI pada penggerak motor BLDC. Penelitian ini membandingkan kinerja kontrol motor BLDC menggunakan PI konvensional dan PI yang dioptimalkan dengan RL. Implementasi metode ini dapat meningkatkan performansi, stabilitas, dan efisiensi kontrol motor BLDC pada berbagai kondisi operasional. Berdasarkan data pengujian pemodelan simulasi rata-rata peningkatan performasi kecepatan berupa overshoot yang berkurang hingga 9,4%, error steady state berkurang sebesar 0,66%, settling time berkurang 0,32s meskipun ada sedikit peningkatan pada rise time sebesar 0.18%. Sedangkan pada pengujian prototipe rata-rata peningkatan performansi berupa berkurangnya overshoot hingga 11,51%, setlling time berkurang sebesar 0,05s dan efisiensi meningkat sebesar 2,96% meskipun pada rise time sedikit mengalami peningkatan sebesar 0,04s dan, serta error staedy state 0,38%. dengan data tersebut dapat di ketahui bahwa optimasi RL pada sistem kontrol kecepatan motor BLDC berbasis FOC cukup efektif.
=================================================================================================================================
BLDC motors are widely used in industrial applications, including electric vehicles and robotics. This is because BLDC motors have high energy efficiency, reliability, and performance. Controlling BLDC motors under dynamic conditions involving changes in speed and beban often faces challenges, especially when using conventional PI control that requires complex manual adjustments. This thesis will develop a control system for Brushless DC (BLDC) motors based on Field-Oriented Control (FOC) using the Proportional-Integral (PI) method optimized with Reinforcement Learning (RL). This thesis proposes the use of RL, specifically the Deep Deterministic Policy Gradient (DDPG) algorithm, to optimize PI control in BLDC motor drives. This study compares the performance of BLDC motor control using conventional PI and PI optimized with RL. The implementation of this method is expected to improve the performance, stability, and efficiency of BLDC motor control under various operational conditions. Based on simulation test data, the average increase in speed performance is in the form of overshoot reduced by 9.4%, stable state error reduced by 0.66%, settling time reduced by 0.32s although there is a slight increase in rise time by 0.18%. While in prototipe testing, the average increase in performance is in the form of a decrease in overshoot up to 11.51%, settling time decreased by 0.05s and efficiency increased by 2.96% even though the rise time slightly increased by 0.04s, as well as a steady state error of 0.38%. with these data it can be seen that RL optimization in the FOC-based BLDC motor speed control system is quite effective.

Item Type:	Thesis (Masters)
Uncontrolled Keywords:	Motor BLDC, FOC, Kontrol PI, Reinforcement Learning, BLDC Motor, FOC, PI Control, Reinforcement Learning
Subjects:	T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK1007 Electric power systems control T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK2681.O85 Electric motors, Brushless.
Divisions:	Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Electrical Engineering > 20101-(S2) Master Thesis
Depositing User:	Prishandy Hamami Amrulloh
Date Deposited:	21 Jan 2026 08:45
Last Modified:	21 Jan 2026 08:45
URI:	http://repository.its.ac.id/id/eprint/130001

Actions (login required)

View Item